Skip to main content
Internet Archive's 25th Anniversary Logo

Full text of "The telephone and telephone exchanges; their invention and development"

See other formats







AND TELEPHONY. By J. A. FLEMING, M.A., D.Sc., F.R.S. With 7 
Plates and numerous other Illustrations. 8vo. 

By J. A. FLEMING, M.A., D.Sc., F.R.S. 8vo. 7*. t>d. net. 

TELEGRAPHY. By Sir W. H. PREECE, K.C.B , F.R.S. ; and Sir 
J. SIVEWRIGHT, M.A., K.C.M.G. Revised and Partly Re-written by 
W. LLEWELLYN PREECE, M.Inst.C.E., Mem.Inst.E.E. With 256 Illustra- 
tions. 8vo. 7.!-. 6d. net. 

STANLEY, B.A., M.I.E.E. With Illustrations. 8vo. Js. 6d. net. 

TRICITY : a Combined Lecture and Laboratory Course. By WILLIAM 
ALLANACH, B.Sc. (Lond. ), Prinicipal of the Technical School, Southport. 
Crown 8vo. 3^. 6d. 

MAGNETISM AND ELECTRICITY. A Manual for Students in 
Advanced Classes. By E. E. BROOKS, B.Sc. (Lond.), A.M.I.E.E., Head of 
the Department of Physics and Electrical Engineering in the Municipal Tech- 
nical School, Leicester, and A. W. POYSER, M.A. (Trinity College, Dublin), 
Head Master of the Grammar School, Wisbech. With 413 Illustrations. 
Crown 8vo. Js. 6d. 

With 1 66 Illustrations. By ARCHIBALD BURSILL, A. M.I.E.E., Lecturer in 
Physics and Electrical Engineering at Paddington Technical Institute. Crown 
8vo. 3*. net. (Longmans 1 Technical Handicraft Series.) 

Consulting Engineer, and Special Lecturer in Electrical Design in the University 
of Manchester ; and CHARLES F. SMITH, M.I.E.E., Assoc.M.Inst.C.E., Whit. 
Schol., Lecturer in Electrical Engineering, the University of Manchester. With 
114 Diagrams. 8vo. Js. 6d. net. 

By A. P. M. FLEMING, M.I.E.E., and R. JOHNSON, \fr. M.I.E.E. With 

Diagrams. Medium 8vo. Js. 6J. net. 

Founded on JOUBERT'S 'Traite Elementaire d'Electricite.' With 376 Illustra- 
tions and Diagrams. 8vo.-ioj. 6d. net. 

suitable for the use of First and Second year Students of Electrical Engineering. 
By W. H. N. JAMES, A.R.C.SC. (Lond.), A.I.E.E., and D. L. SANDS. Crown 
8vo. 3-r. bd. net. 

HENDERSON, D.Sc., F.R.S.E. 157 Illustrations and Diagrams. Crown 8vo. 
js. 6d. 

LONGMANS, GREEN, & CO. 39 Paternoster Row, London, 

New York, Bombay, Calcutta, and Madras. 


JOYCE, A.I.E.E. Crown 8vo. 5*. 

By J. L. LA COUR, Vesteras, and O. S. BRAGSTAD, Trondhjem. Translated 
by STANLEY P. SMITH, D.Sc., A.M.InsLC.E., A.M.I.E.E. With Diagrams, 
8vo. (Elements of Electrical Engineering.) 

ING, being a Compilation of the Questions set by the City and Guilds of London 
Institute in the Preliminary Grade of Electric Lighting, from 1899 to 1907 
inclusive, with solutions to all Questions. By A. E. MOORE, A.M.I.E.E., and 
FRANK SHAW. Crown 8vo. 23. 6ef. 

(ist Hons. ), late Rew Research Scholar, Liverpool ; Lecturer in Electrical 
Engineering, Royal Technical Institute, Salford. With Diagrams. Crown 8vo. 
2s. 6d. net. 

M.I.K.E., A.M.I.Mech.E. With 231 Illustrations. 8vo. 8s. 6d. 


POYSER, M.A. With 235 Illustrations. Crown 8vo. 2s. 6d. 

W. POYSER, M.A. With 317 Illustrations. Crown 8vo. 4^. 6d. 

By JOHN II. MORECROFT, E.E., Member A.I.E.E. A sistant Professor of 
Electrical Engineering, Columbia University. 8vo. 7^. (ju. net. 

RHODES, D.Sc., Head of the Electrical Engineering Department, The Royal 
Technical Institute, Salford. With Diagrams. Crown 8vo. 2s. 6d. net. 
(Longmans' technical Handicraft Series.) 

POWER, ARTISANS AND STUDENTS (embracing thse branches pre- 
scribed in the Syllabus issued by the City and Guilds Technical Institute). By 
W. SLINGO, A. HKOOKER, Engineer-in-Chief of the General Post Office, and 
T. F. WALL, M.Sc. With 389 Illustrations. Crown 8vo. I2s. 6d. 

Chief of the General Post Office. With 98 Illustrations. Crown 8vo. 2s. 


Vice-President of the Institution of Electrical Engineers. With 17 Folding 
Plates, and 186 Illustrations. 8vo. 21*. net. (Longmans' Electrical Engineer- 
ing Series.) 

KI.I.CTRICITY AND MAGNETISM. For Advanced Students. 
l',y SYDNKV G. STARLING, B.Sc., A.R.C.SC., Head of the Physical Department 
in the Wist Hum Municipal Technical Institute. With Diagrams. Crown 
8vo. 7*. 6J. net. 

I.<>.\i,M.\\S, GREEN, & CO. 39 Paternoster Row, London, 

New York, Bombay, Calcutta, and Madras. 

















All rights reserved 


IN the following pages the attempt has been made so to relate the 
inventions and developments in the telephone field that the record 
may constitute in effect a short history of the telephone industry. 
The original intention was to write a history, but I was advised 
on authority which in the book world is entitled to respect not to 
exceed the limits of a single volume. As that advice has been 
followed it is considered that the word ' history ' as a title should 
be held in reserve for the use of the author or authors who may 
produce the comprehensive treatise which the subject deserves. 

The original intention has, however, been followed in an abbre- 
viated form. The principal inventions have been selected. The 
circumstances leading up to them, the developments resulting 
from them, and the influences bearing on them, have been con- 
sidered with all the detail that space permits. But telephone 
exchange service is not merely a matter of plant and inven- 
tion. Technical, commercial, and political threads compose the 
fabric, and they are interwoven in the record. The recital of the 
prior causes has served in most cases to make clear the technical 

In recounting the progress of Bell's invention reference is fre- 
quently made to the evidence given by him in various suits, noted 
for the sake of brevity as ' Deposition.' The full title is ' The Bell 
Telephone : The Deposition of Alexander Graham Bell in the suit 
brought by the United States to annul the Bell patents.' The suit 
was abandoned, but the deposition (in which is included evidence 
taken in other suits) was printed in 1908 by the American Bell 
Telephone Co. ' because of its historical value and scientific interest.' 
This book of 469 pages, though not published in the ordinary sense, 
is available for reference in most technical libraries. 


Some information regarding early exchange service has been 
taken from the ' Boston Electrical Handbook/ which was prepared 
under the auspices of the American Institute of Electrical Engineers 
as ' a guide for visitors from abroad attending the International 
Electrical Congress, St. Louis, Mo., September 1904.' This work 
though printed only for such private circulation may also be found 
in most technical libraries. 

More frequent references are made to the reports of the National 
Telephone Exchange Association. The word ' National ' has been 
adopted frequently in connection with the telephone there was, 
for example, the National Bell Co. in the United States and the 
National Telephone Co. in Great Britain so that it may be well 
to state here that the National Telephone Exchange Association 
was an association of Telephone Companies in the United States 
who were licensees under the Bell patents. The reports of the 
Association constitute a most valuable record of the annual 
sometimes . semi-annual discussions of the active participants in 
the early developments of a new industry. 

After the first decade progress was less represented in publica- 
tions. Committees of experts discussed the problems and originated 
improvements, but their reports were not published. Descriptions 
of the apparatus eventually put into use are readily available, but 
they do not indicate what I was desirous of including the evolu- 
tionary stages of progress in each section. In order to obtain 
authoritative information on these points I applied to the Pre- 
sident of the Western Electric Co. for permission to examine and 
make extracts from the Reports of the Committees and Conferences 
of experts of that Company and of the American Telephone Com- 
panies who considered from time to time the changes desirable 
in the art. The obligation which I am under to Mr. H. B. Thayer 
for kindly placing those voluminous documents at my disposal 
will be shared by the reader. The more important developments 
in Switchboards, Cables, and Exchange service generally are here 
i from the records of the deliberations of those experts, and 
the information is now published for the first time. 

My obligations to my friend Mr. T. 1). Lockwood cannot be con- 
i- ly Dialed. Some are mentioned in the course of the book, but I 
am also indebted to him for the loan of documents from which much 
of the information regarding the early applications in the United 


States related in Chapters VIII. and IX. is derived ; for practical 
help in solving doubts or difficulties in some dates and facts ; and 
for much appreciated encouragement in the prosecution of the 

My indebtedness to the Institution of Electrical Engineers and 
the American Institute of Electrical Engineers will be obvious by 
the references to their respective publications. 

My thanks are due also to the Editors of various other publica- 
tions for permission, readily and cordially given, to reproduce 
illustrations, the sources of which are duly noted in their respective 
places. In general, these illustrations are limited to early examples 
of historic interest. Central office interiors, for instance, of later 
date are well known to telephone readers and easily accessible to 

I am indebted to my son for assistance in the preparation of the 
illustrations for the press. 



October 1915. 



I. INTRODUCTORY ....... i 

II. THE SPOKEN WORD . . . . . -14 








XI. THE MICROPHONE . . . . . . .113 


XIII. CALL BELLS ....... 140 



UNITED STATES . . . . . . .179 



ABROAD ........ 191 






CONFERENCES OF 1887, 1889,., AND 1891) . . 279 










XXIX. INSTRUMENTS ....... 446 




XXXIII. CONCLUSION . . . . . . -517 









THE telephone to which this book relates is the electric speaking 
telephone that came into the world at the beginning of the last 
quarter of the nineteenth century. 

The name itself is much older than the instrument to which 
it is now exclusively applied, for before the advent of the electric 
speaking telephone there were appliances called telephones through 
which words were spoken though these appliances were not operated 
electrically, and there were telephones operated electrically which 
were not available for conversation. Acoustic telegraphs have 
been called telephones, and, on the other hand, telephones have 
been designated speaking telegraphs. 

The efforts to extend the distance to which speech might be 
transmitted were, in earlier times, limited to the use of a form of 
trumpet to increase the power of transmission or reception the 
sound waves through the intervening space being subject to the 
attenuating effect of the atmosphere or of a tube between speaker 
and listener, within which tube the sound waves were confined. 

A specimen of the ear trumpet was exhibited at the Royal 
Society in 1668, and of it a contemporary writes : 

I did try the use of the Otacousticon, which was only a great 
glass bottle broke at the bottom, putting the neck to my ear, and 
there I did plainly hear the dancing of the oars of the boats in the 
Thames to Arundel Gallery window, which, without it ; I could not 
in the least do. 1 

1 Diary of Samuel Pepys, April 2, 1668. 


The speaking trumpet was developed by Sir Samuel Morland, 
Master of Mechanics to Charles II. Morland published an elaborate 
treatise on the subject. The following is a literal transcription of 
the title page : 

Tuba Stentoro-Phonica an Instrument of Excellent Use as well 
at sea as at Land ; Invented and variously experimented in the 
year 1670 and Humbly Presented to the King's most Excellent 
Majesty Charles II in the year 1671 By S. Morland. The Instru- 
ments (or speaking Trumpets) of all Sizes and Dimensions, are made 
and Sold by M r - Simon Beal, one of His Majesties Trump 5 in 
Suffolk Street. London. Printed by W. Godbid and are to be sold 
by M. Pitt at the White Hart in Little Britain, 1671. 

The work describes and illustrates a number of trumpets ; con- 
tains the copy of a letter from Fr. Digby to Lord Arlington, Principal 
Secretary of State, which shows that the King took great interest 
in the trumpets since the letter was written ' as the best way of 
satisfying His Majesty concerning them ' ; and includes also ' A 
short discourse Touching the Nature of Sounds, and the manner 
how (as I conceive) they are magnified, or rather multiplied, by the 
Tuba Stentoro-Phonica.' 

The practical results are recorded in the course of a contem- 
porary review or abstract, such as might be found in Science 
Abstracts of the present day, which appeared in the Philo- 
sophical Transactions : 

The Author of this instrument relates first the several trials made 
with it, of which the most considerable was, that the largest of those 
that have been as yet employed turned trumpet-wise being 5 feet 
6 inches long, and of 21 inches diameter at the great end, and 2 inches 
at the less ; when by His Majesty's special command it was tried 
at Deal Castle by the Governor thereof, the voice was plainly 
heard off at Sea as far as the King's Ships usually ride, which is 
brtwrm two and three miles, at a time when the wind blew from 
the shore. 1 

It is to Morland's speaking trumpet that Samuel Butler refers 
in the third part of ' Hud bras,' which was published in 1678 : 

I heard a formidable noise, 
Loud as the stent'rophonic voice, 
That roared far off.* 

Tin re was controversy over the claim to the invention of the 
^1>(. iking trumpet as there was in after years over the invention 
of the speaking telephone. Butler elsewhere describes it as a 

1 Philosophical Transactions, Xo. 79, p. 3056. Hudibras, III. 1. 251. 


'new nicked-namcd old invention.' 1 But the contention mainly 
lay between Morland and Kircher. Dr. Derham (1657-1735), 
an eminent English divine and natural philosopher, gave the 
credit to the latter. 

A century later the evidence was carefully considered by 
Professor Beckmann of the University of Gottingen in his ' History 
of Inventions and Discoveries translated by William Johnston 
and published by J. Bell No. 148 Oxford Street, London ' in 1797. 
Beckmann relates that ' Derham refuses the invention to his 
countryman and gives it to Kircher.' But Beckmann returns 
the compliment. He refuses the invention to his own countryman 
and gives it to Morland. 

Kircher [he says] pretends that so early as the year 1649 he had 
caused such a machine to be set up in the Jesuits College. But 
supposing this to be true, it can only be said that he then approached 
very near to the invention of the speaking trumpet, by an instru- 
ment however which in reality was calculated to strengthen the 
hearing and not the voice, and therefore only the half is true of 
what he advanced in his preface in 1673, that twenty years before 
he had described in his Musurgia the trumpet invented in England. 2 

An important distinction is thus drawn between a speaking 
trumpet intended to strengthen the sounds transmitted and an 
ear trumpet for increasing the audibility of sounds received. Beck- 
mann was disposed to give Kircher credit for the invention of a 
receiver and to deny him the merit of inventing a transmitter. 
But the judicial summing-up and verdict on the rival claims have 
their principal interest in the emphasis laid on the test of application 
to practical utility : 

When I unite all the evidence in favour of Kircher it appears to be 
certain that he made known and employed the ear trumpet earlier 
than the portable speaking trumpet ; that he, however, approached 
very near to the invention of the latter, but did not cause one to be 
constructed before Sir [Samuel] Morland, to whom the honour 
belongs of having first brought it to that state as to be of real use. 3 

However efficacious the speaking trumpet may have been for 
the use of His Majesty's navy in the seventeenth century, it was 
not equal to the requirements of the service in the nineteenth. 
Flags and semaphores had a longer reach, though the need of codes 
made them less eloquent. Before he became a writer of matchless 
fiction Captain Marryat was ' the inventor (1817) of the completest 

1 An Heroical Epistle of Hudibras to Sidrophel, 1. 21. 

2 History of Inventions, i. 161. 3 Ibid. i. 164. 

B 2 


code, of signals ever introduced,' which was in use in the royal 
and mercantile service of Great Britain as well as in navies of 
foreign nations. But, perfect as a code might be, flags and sema- 
phores were only able to transmit its meanings in a clear 

In 1845 another captain, one John Taylor, invented an instru- 
ment ' for conveying signals during foggy weather by sounds pro- 
duced by means of compressed air forced through trumpets.' This 
instrument was called ' The Telephone,' and is briefly described 
in the ' Year Book of Facts in Science and Art,' 1845, p. 55. 

The same publication in 1854 (p. 130) describes a ' New 
Telephone.' This was the invention of M. Sudre of Paris, and did 
not vary materially in principle from that of Taylor ; but M. Sudre 
had probably developed a more complete code. There were three 
notes, and in a practical demonstration, a sentence having been 
written down, the notes were struck by M. Sudre 

alternately, according to his method, when a third person, 
without any previous knowledge of the writing, repeated the words 
merely from hearing the notes.' 

The French Institute and other scientific bodies were said to have 
passed ' high encomiums ' upon this invention. 

In the London Exhibition of 1851 Francis Whishaw of 9 John 
Street, Adelphi, exhibited a ' Gutta Percha Telephone,' l which 
is not described, but the prior use of the word ' telephone ' in 
connection with instruments of the trumpet type justifies the 
inference that it was a speaking trumpet. The same manu- 
facturer exhibited the telekouphonon, which is described and 
illustrated. It is the familiar domestic speaking tube, ' or speaking 
telegraph, consisting of gutta percha, glass, metal, or other proper 
tubing. 1 In an article on gutta percha and its manufactures in 
' Stories of Inventors and Discoverers,' 1860 (p. 328), it is asserted 
that Mr. Whishaw had 

early discovered the valuable property which Gutta Percha possesses 
for the conveyance of sound, and accordingly made of it the Tela- 
kouphanon or speaking trumpet through which, by simply 
whispering, the voice could be audibly conducted for a distance 
of three quarters of a mile, and a conversation by this means be 
kept up. 

The author has not correctly transcribed the name, but is not so 
far out as the Jury whose Report has telekerephona. 2 

1 Official Description and Illustrated Catalogue of the Great Exhibition, 
1851, i. 455. * Reports of the Juries, p. 598. 


Since the telekouphonon, as the catalogue has it, was a speaking 
tube for domestic use and not a speaking trumpet, it is probable 
that the name was incorrectly introduced into the article quoted. 
What was probably intended to be referred to was a gutta percha 
speaking trumpet, and this was catalogued as a ' Gutta percha 

Though applied to signalling trumpets the word ' telephone ' 
does not appear to have come into use in England in connection 
with speaking tubes as it had in Germany. The examiner in the 
United States Patent Office gives two references of the year 1869 
and one of 1871 in proof that ' the ordinary speaking tube was 
known in Germany as the " Telephone." ' l 

The Didaskalia, published at Frankfort, in its issue of 
September 28, 1854, referred to Bourseul's suggestion 2 for the 
electrical transmission of speech under the heading of ' Electrische 
Telephonic,' and Reis applied the name of The Telephone to 
the apparatus which he devised in i86i. 3 The still earlier use of 
the word ' telephone ' in connection with apparatus devised by 
Wheatstone is referred to later in this chapter. 

While the speaking trumpet is given a section to itself in 
Beckmann's history, speaking machines are ominously included in 
a chapter entitled ' Jugglers.' The early machines of this kind were 
either so constructed as to conceal a confederate or were devices 
connected by a speaking tube with a distant point where such 
confederate was more securely hidden. The latter was the plan 
generally adopted when these deceptions were practised for religious 

W T hether [says Beckmann] the head of Orpheus spoke in the 
island of Lesbos, or, what is more probable, the answers were con- 
veyed to it by the priests, as was the case with the tripod at Delphi, 
cannot with certainty be determined. That the impostor Alexander, 
however, caused his .^sculapius to speak in this manner is expressly 
related by Lucian. 4 

Those speaking machines which purported to answer various 
questions submitted to them were comprised within a figure or 
sometimes only a head placed upon a box, the front of which ' for 
the better deception ' was filled with ' a pair of bellows, a sounding 
board, cylinder and pipes supposed to represent the organs of 
speech.' The popular opinion of these machines varied. Some 
affirmed that the voice issued from the machine, others that the 
juggler himself answered by speaking as ventriloquists do, and 

1 The Speaking Telephone Interferences, p. 4. 

2 Applications del' Electricity, Du Moncel, 1854. 

3 See Chapter XII. * History of Inventions, iii. 333. 


some believed that the answers were given by a man somewhere 
concealed. When the illusion was detected the populace imagined 
they had a right to avenge themselves for being imposed on, which 
the historian evidently regarded as being very unreasonable. He 
says : 

For my part I do not see why a juggler, with a speaking 
machine, is a more culpable impostor than he who pretends 
to breathe out flames and to swallow boiling oil, or to make 
puppets speak, as in the Chinese shadows. The spectators pay for 
the pleasure which they receive from a well-concealed deception, 
and with greater satisfaction the more difficult it is to discover it. 1 

In the eighteenth century much interest was taken in automata. 
Chess players, singing birds, and other mechanical contrivances 
were constructed for the edification or amusement of European 
courts. Useless in themselves, they probably tended to develop 
the skill of the mechanicians, whose aspirations rose from chess 
players and singing birds to speaking machines. Towards the end 
of the century the scientific interest in speech increased, and it is to 
the Imperial Academy of St. Petersburg (or Petrograd) that credit 
has to be given for the first substantial encouragement to inquiry. 
In the year 1779 the Academy proposed for the annual prize two 
questions. Firstly, what was the nature and character of the vowels 
a e i o u, ' so different from each other ' ; and secondly, could an 
instrument be constructed like the vox humana pipes of the organ, 
which should accurately express the sounds of the vowels ? The 
prize was awarded to Professor Kratzenstein, who 

constructed a series of tubes, which, when applied to an organ 
bellows, imitated with tolerable accuracy the five vowel sounds 
required. These tubes were of the most grotesque and complicated 
forms ; for which no reason was offered except that experience 
had shown these forms to be the best adapted to the production of 
the sounds in question. Some of the pipes were rendered vocal 
by the application of vibrating reeds, like those in clarionets ; 
while others were open in the manner of a common organ pipe. 2 

In 1791 De Kempelen of Vienna published a description of his 
talking machine in which the bellows and pipes were not supplied 
for the purpose of deception, but, like those of Kratzenstein, \vriv 
really used. De Kempelen attempted to imitate not only vowel 
sounds but also consonants. 

Mr. Willis of Cambridge (1829) was less ambitious but more 

1 History of Inventions, iii. 331. 

1 Saturday Magazine, February n, 1843. \Vheatstone (Scientific Papers, 
p. 352) gives a more abbreviated account, but apparently from the same source. 


thorough. He limited his efforts to vowel sounds and formed the 
foundation of the scientific inquiry which Wheatstone and Helm- 
holtz continued. A contemporary of Wheatstone's in the analysis 
of speech sounds was Dr. Rush of Philadelphia, who wrote a work 
entitled ' The Philosophy of the Human Voice.' 

Mr. Reale exhibited before the American Philosophical Society 
in 1843 a machine capable of enunciating various letters and words. 
It is said that the instrument took him sixteen years to construct, 
and that he destroyed it ' in a frenzy,' presumably of despair ; and 
Professor Faber of Vienna produced a machine, considered to be 
more successful than any of its predecessors, which was exhibited 
in London in 1846. 

In all the machines of this type the designers sought to produce 
speech. They succeeded only within a strictly limited range, 
and very imperfectly even then. The so-called speaking machines 
which spoke were really speaking tubes with a talker at one end 
and a listener at the other, the conducting medium being air, the 
waves which reached the ear being the waves directly produced by 
the voice, but retaining their power to a greater distance than in the 
atmosphere because the walls of the tube reduced the loss of energy 
from spherical action. 

The utilisation of a solid medium for transmission of sound to 
a distance was developed by Wheatstone, who, in the introduction 
to his paper ' On the transmission of musical sounds through solid 
linear conductors, and on their subsequent reciprocation,' published 
in the ' Journal of the Royal Institution,' 1831, vol. 2, wrote : 

The fact of the transmission of sound through solid bodies, as 
when a stick or a metal rod is placed with one extremity to the ear, 
and is struck or scratched at the other end, did not escape the 
observations of the ancient philosophers ; but it was for a long time 
erroneously supposed that an aeriform medium was alone capable 
of receiving sonorous impressions ; and in conformity with this 
opinion, Lord Bacon, when noticing this experiment, assumes that 
the sound is propagated by spirits contained within the pores of 
the body. The first correct observations on this subject appear 
to have been made by Dr. Hooke in 1667, who made an experiment 
with a distended wire of sufficient length to observe that the same 
sound was propagated far swifter through the wire than through 
the air. Professor Wunsch, of Berlin, made, in 1788, a similar 
experiment, substituting 1728 feet of connected wooden laths for 
the wire, and confirmed Dr. Hooke's results. 1 

Beckmann makes more definite mention of the ancient philo- 
sophers in connection with this subject. ' It had been remarked,' he 

1 Wheatstone's Scientific Papers, p. 47. 


says, ' even in Pliny's time, that the least touching of a beam of wood 
could be heard when one placed one's ear at the other end.' l 

The year 1667 given by Wheatstone as that of Dr. Hooke's 
observations has been adopted generally by later writers. In a 
note to his paper Wheatstone quotes the Preface to Hooke's 
' Micrographia.' The edition of this work published in 1667 was the 
second. The first appeared in 1665, having been ' ordered to be 
printed by the Council of the Royal Society, November 23, 1664.' 
The paragraph partly quoted by Wheatstone appears in the first 
edition (1665) on the 'eighth page of the Preface, the complete 
paragraph being as follows : 

And as Glasses have highly promoted our seeing, so 'tis not 
improbable, but that there may be found many Mechanical In- 
ventions to improve our other Senses, of hearing, smelling, tasting, 
touching. Tis not impossible to hear a whisper a furlong's distance, 
it having been already done ; and perhaps the nature of the thing 
would not make it more impossible, though that furlong should 
be ten times multiply 'd. And though some famous Authors have 
affirm'd it impossible to hear through the thinnest plate of Muscovy- 
glass ; yet I know a way, by which 'tis easie enough to hear one 
speak through a wall a yard thick. It has not yet been thoroughly 
examin'd, how far Otocousticons may be improv'd, nor what 
other wayes there may be of quickning our hearing, or conveying 
sound through other bodies then the Air : for that that is not the 
only medium, I can assure the Reader, that I have, by the help 
of a distended wire, propagated the sound to a very considerable 
distance in an instant, or with as seemingly quick a motion as that 
of light, at least, incomparably swifter then that, which at the 
same time was propagated through the Air ; and this not only in a 
straight line, or direct, but in one bended in many angles. 

Ten years before writing his Royal Institution paper Charles 
Wheatstone, in 1821, whilst an assistant to his uncle, a musical 
instrument maker, devised a method of transmitting music by 
means of wooden rods. The rod at the transmitting end was 
attached to a piano and at the receiving end to a sounding board 
designed to represent an ancient lyre, the conventional repre- 
sentative of music in art and literature. The device attracted 
considerable attention in scientific circles and amongst the public. 
Accounts of the performance were given in the Repository of Arts, 
September i, and the Literary Gazette, September 15, 1821, but 
the principle of operation was not disclosed. In his paper ' New 
i i incuts on Sound,' published in Thomson's ' Annals of 
Philosophy,' iSj3, Whcatstone described it fully. 

1 History of Inventions, i. 152. 


In my first experiments on this subject [he says] I placed a 
tuning fork or a chord extended on a bow, on the extremity of 
a glass or metallic rod five feet in length, communicating with 
a sounding board ; the sound was heard as instantaneously as when 
the fork was in immediate contact ; and it immediately ceased 
when the rod was removed from the sounding board or the fork 
from the rod. From this it is evident that the vibrations, inaudible 
in their transmission, being multiplied by meeting with a sonorous 
body, become very sensibly heard. Pursuing my investigations 
on this subject, I have discovered means for transmitting, through 
rods of much greater lengths and of very considerable thicknesses, 
the sounds of all musical instruments dependent on the vibrations 
of solid bodies, and of many descriptions of wind instruments. 
It is astonishing how all the varieties of tune, quality, and audi- 
bility, and all the combinations of harmony, are thus transmitted 
unimpaired, and again rendered audible by communication with an 
appropriate receiver. One of the practical applications of this 
discovery has been exhibited in London about two years, under the 
appellation of ' The Enchanted Lyre.' So perfect was the illusion 
in this instance from the intense vibratory state of the reciprocating 
instrument, and from the interception of the sounds of the distant 
exciting one, that it was universally imagined to be one of the 
highest efforts of ingenuity in musical mechanism. 1 

That Wheatstone called the enchanted lyre a telephone is 
a statement which is to be found in most of the books on the subject, 
but, so far as I have been able to examine them, without any 
reference to source or authority. The inference has very generally 
been drawn that the enchanted lyre was called a telephone from 
its inception ; which if correct would carry the word back to 1821. 
But there were good reasons why no such descriptive name should 
have been given to it then. 

A certain degree of mystery was necessary in order to arouse 
public interest. Wheatstone used to go through the form of 
winding up the lyre with a key, but neither of the magazine writers 
was misled thereby. The winding up was regarded as ' evidently 
a mere ruse.' Moreover, the Repository of Arts thought ' proper to 
add ' the statement of Mr. Wheatstone that the exhibition was 
' the application of a general principle for conducting sound, which 
principle he professed himself to be capable of carrying to a much 
greater extent.' The mild deception practised was akin to that 
of the conjurer who excites interest by mystification, but it serves 
to show that it was not then Wheatstone's desire to explain so 
much as would have been explained if he had applied the word 
' telephone ' to the contrivance. The name of ' The Enchanted 
Lyre ' was too simple to stand alone and needed supplementing 

1 \Vheatstone's Scientific Papers, p. 7. 


by a compound from a dead language. This alternative name was 
' The Acoucryptophone, ' l in which the mysterious or secret feature 
was comprised. But not only in this early stage did Wheatstone 
refrain from calling the apparatus a telephone. Neither in 
Thomson's ' Annals of Philosophy ' of 1823, where the principle of 
operation was first disclosed, nor in any subsequent description 
included in his collected Papers does Wheatstone use the word 
' telephone.' It is just such a word as he might have coined, and 
probably did coin, but whether before the electric telegraph had 
come into operation is doubtful. 

Wheatstone collaborated with Cooke in perfecting and intro- 
ducing their electric telegraph in 1837. Three years later, in conse- 
quence of a misunderstanding which had prevailed respecting their 
' relative positions in connection with the invention,' the arbitration 
of Sir M. Isambard Brunei and Professor J. F. Daniel was sought. 
In the course of this arbitration, which was agreed upon on 
November 16, 1840, the word 'telephone' is frequently used, and 
this is probably its earliest use. 

In a printed pamphlet constituting his case Professor Wheatstone 
says : 

The subject of telegraphic communication has for a long series 
of years occupied my thoughts. When I made in 1823 the discovery 
that sounds of all kinds might be transmitted perfectly and power- 
fully through solid wires and rods, and might be reproduced in 
distant places, I thought that I had an efficient and economical 
means of establishing a telegraphic (or rather a telephonic) com- 
munication between two distant places. 2 

In the same ' case ' he refers to his proposal for a telegraph 
with two wires, in which electric sparks and his revolving mirror 
were employed. He describes this briefly, and adds : 

I have not continued these experiments, but the principles 
employed in them have been of great use to me in my most recent 
investigations. My rhythmical telephone, invented last year [1839], 
is but a modification of it, in which the strokes of a bell are substi- 
tuted for the sparks, and the voltaic current for the electric 

While Wheatstone himself uses the word ' telephone ' only 
in connection with an electric bell, and the word ' telephonic ' ' in 

1 Min. Proc. Institution of Civil Engineers, xlvii. 284. 

The Electric Telegraph was it invented by Prof. Wheatstone, Part II. 
p. 81. * Ibid. p. 84. 

4 The Imperial Dictionary published in 1854 contains the word 
' telephonic,' but not the word ' telephone.' 


a parenthetic way for the purpose of verbal accuracy, his 
opponents in the arbitration proceedings frequently describe 
his apparatus as the telephone. In the address of Mr. Wilson laid 
before the arbitrators on February 27, 1841, as an introduction to 
the evidence to be adduced on Mr. Cooke's behalf, repeated reference 
is made to ' the telephone ' : 

I maintain, gentlemen, that it was then [after Cooke's introduc- 
tion to him] and not till then, that Professor Wheatstone became 
connected with the Practical Electric Telegraph ; and that up to 
that time he had done nothing in any respect more practical than 
his telegraph with common electricity and a revolving mirror, or 
his idea, to which his letter refers, of a telephone between London 
and Edinburgh. 1 

It is evident that Professor Wheatstone had turned his mind to 
the Telephone and to the Electric Telegraph for fourteen or fifteen 
years without any practical result. 2 

There are other references of a similar nature, but further 
quotation is unnecessary. The word ' telephone ' is here used in 
1841 not by Wheatstone himself but by the legal representative 
of the opposing party, and it was used with greater frequency 
than would seem to have been really necessary. The reason must 
be sought in the nature of the inquiry submitted to arbitration : 
To whom was due the credit for the introduction of the Practical 
Electric Telegraph ? The frequent use of the word ' telephone ' 
by Mr. Cooke's solicitor is probably to be explained by his desire 
to emphasise the unpractical nature of Wheatstone's idea. ' The 
telephone ' was in fact used as a sort of term of reproach. It 
represented something unpractical leading to nothing, whilst the 
electric telegraph, on the other hand, was the ' Practical Electric 
Telegraph,' and Cooke's work thereon had resulted in great public 

This line of argument in view of the purpose of the arbitration 
was not unfair. Valuable as it was scientifically, the enchanted 
lyre experiment had never been carried practically beyond the 
room-to-room stage. But whatever may have been the prior use, 
as to which some uncertainty may be felt, after this period the 
term ' telephone ' was very generally applied to it. 

In 1855 there was a demonstration at the Polytechnic Institution 
before Queen Victoria, which was referred to by Mr. C. K. Salaman 
in a letter to the ' Choir ' as a ' telephone concert,' 3 and an illustra- 
tion accompanying a lecture-table modification of this device is 

1 The Electric Telegraph was it invented by Prof. Wheatstone, Part II. 
p. 121. 

2 Ibid. p. 124. 3 Post and Telegraphs, Tegg, 1878, p. 290. 


entitled ' The Miniature Telephonic Concert ' in Pepper's 
' Cyclopaedic Science Simplified/ published in iSGg. 1 Tyndall 
utilised the illustration in the course of a series of lectures on 
Sound at the Society of Arts. The Society's Journal of May 5, 
iSyi, 2 states that ' among many other interesting illustrations was 
an example of the " telephone concert " of Sir Charles Wheatstone.' 
And in the ' Handbook to the Special Loan Collection of Scientific 
Apparatus/ 1876? it is said that the transmission of sound in wood 
' was ingeniously demonstrated by Wheatstone. His telephone 
consisted of long rods of light pine/ etc. 

Wheatstone's suggestion had nothing in common with the 
telephone now in general use. He aimed at the mechanical or 
molecular transmission of acoustic vibrations, and even in that 
(except for purely local demonstration) had not gone beyond 
the stage of suggestion. It is from the theoretical, rather than the 
practical, standpoint that we should consider the suggestion that 
Mr. Cooke's solicitor called a telephone. 

The conclusion of Wheatstone's paper, of which the introduc- 
tory passage was quoted on page 7, is as follows : 

When sound is allowed to diffuse itself in all directions as from a 
centre, its intensity, according to theory, decreases as the square of 
the distance increases ; but if it be confined to one rectilinear 
direction, no diminution of intensity ought to take place. But this 
is on the supposition that the conducting body possesses perfect 
homogeneity, and is uniform in its structure, conditions which never 
obtain in our actual experiments. Could any conducting substance 
be rendered perfectly equal in density and elasticity so as to allow 
the undulations to proceed with a uniform velocity without any 
reflections or interferences, it would be as easy to transmit sounds 
through such conductors from Aberdeen to London as it is now 
to establish a communication from one chamber to another. 
\\ hrther any substance can be rendered thus homogeneous and 
uniform remains for future philosophers to determine. 4 

Wheatstone saw clearly the limitations of any molecular method 
of transmission, and equally clearly the benefits which must result 
from speech transmission if a practicable method could be found. 
He continues : 

The transmission to distant places, and the multiplication of 
musical performances, are objects of far less importance than the 
conveyance of the articulations of speech. I have found by experi- 
ment that all these articulations, as well as the musical inflexions 

1 P. 525- 

* Journal Society of Arts, xix. 510. 3 Handbook, etc., p. 99. 

4 Wheatstone's Scientific Papers, p. 62. 


of the voice, may be perfectly, though feebly transmitted to any of 
the previously described reciprocating instruments by connecting 
the conductor, either immediately with some part of the neck or 
head contiguous to the larynx, or with the sounding board to which 
the mouth of the speaker or singer is closely applied. The almost 
hopeless difficulty of communicating sounds produced in air with 
sufficient intensity to solid bodies might induce us to despair of 
further success ; but could articulations similar to those enounced 
by the human organs of speech be produced immediately in solid 
bodies, their transmission might be effected with any required 
degree of intensity. Some recent investigations lead us to hope 
that we are not far from effecting these desiderata ; and if all the 
articulations were once thus obtained, the construction of a machine 
for the arrangement of them into syllables, words, and sentences 
would demand no knowledge beyond that we already possess. 1 

We are left in doubt as to the nature of the machine to be 
constructed. The ' previously described reciprocating instruments ' 
were of the sounding board or lyre type, comparatively simple in 
construction, but something much more complicated is suggested 
by a machine whose function it should be to arrange the trans- 
mitted vibrations ' into syllables, words, and sentences.' 

It seems to have been assumed that for the reproduction of 
speech, complex contrivances capable of artificially producing 
speech would be required. The suggestion of Bourseul (1854) 
and the experimental efforts of Reis (1861) did but little to dissipate 
the idea. If the question were asked who, amongst the pre- 
decessors of the real inventor, should have produced the telephone 
the unhesitating answer would be Charles Wheatstone. He 
was the earliest practical experimenter in this direction. His 
studies in acoustics were original and profound. He was the first 
to enunciate the theory of vowel tones. He was equally well 
equipped in his knowledge of the science of electricity and of its 
practical applications. He was renowned for his ingenious 
mechanical inventions, and was provided with a staff of workers 
specially trained to produce, in the form of highly finished instru- 
ments, the creations of his brain. Yet the idea which his paper of 
1831 conveys, that some elaborate machines were essential for speech 
transmission, was not withdrawn. Notwithstanding Wheatstone's 
sanguine views regarding the possibility of producing such machines, 
no surprise is felt that inventors were deterred from entering upon 
an enterprise in which the experiments must be costly and the 
expectations of success but slight. Even in the circles of science 
there was still a mystery and a magic in human speech. 

1 \\ heatstone's Scientific Papers, p. 62. 



THE magic and the mystery of the spoken word have cast a spell 
over mankind from the earliest times through all the ages ; the 
greatest homage being paid perhaps by the philosophers of 
Hermopolis, who, materialising the intangible, deified speech and 
attributed to it the power of a Creator. To them there was a very 
real meaning in the phrase ' In the beginning was the Word.' To 
them it was ' speech, and above all the simple emission of the 
voice which gave the world the form it now bears.' 

The magic and the mystery of speech have appealed not the 
less powerfully to those who have regarded it simply as a means 
of transferring thought, of communicating intelligence, or of stirring 
the hearts of men. Whether it were the exhortations of the priests, 
the appeals of the orators of ancient times, or the addresses of more 
recent politicians, the medium of communication was the same 
and the power was the power of speech. 

The stirring times of the nineteenth century revived the power 
of the platform which had been on the wane 1 and gave a new interest 
to the cultivation of the powers of speech, until, as the century 
advanced, professors of elocution were numerous. Some of them 
published books, and the impression which most of these books 
produce is that in the main these teachers of oratory devoted little 
attention to the mystery and mechanism of speech, but dwelt 

1 'It is somewhat hard to realise in the present day, when the platform 
is so great a power in the land, how completely non-existent it was as a factor 
in political life, as the nineteenth century dawned upon the country. Events 
of world-vide importance were occurring abroad ; mighty movements wen- 
beginning at home ; but the Government had closed the avenues to discussion, 
and the people were compelled to silence. Matter enough had they for 
thought ; problems enough to perplex them ; prievances and suffering 
enough to make them cry out ; but the articulate voice came not, was not 
permitted to come ; and though the Press, trammelled and terrorised, acted 
to some extent as a vehicle for the expression of the thoughts of the people, 
the public voice as spoken from the platform was dumb.' The Platform, 
Jephson, i. 297 (Macnullan. 1892). 



much on the effects of gesture and of emphasis. Alexander Bell 
of 25 Norton Street, Portland Place, London, was a professor of 
elocution who published in 1835 ' The Practical Elocutionist/ 
which he dedicated to Lord Brougham. He subsequently published 
other works, all of an elocutionary character rather than analytical 
of speech. 

His son Alexander Melville Bell followed him in the same pro- 
fession, practising in Edinburgh. Alexander Melville Bell made a 
more complete study of the mechanism of speech than any professor 
of elocution before him. From Edinburgh he removed to London, 
and was lecturer on elocution in University College. He published 
works on ' The Principles of Speech, and Cure of Stammering,' 
' Letters and Sounds An Introduction to English Reading on an 
entirely new plan,' ' Observations on Stammering and the Principles 
of Elocution,' and other works of a similar character ; but internal 
evidence shows that in the view of the author his magnum opus 
was ' Visible Speech, the Science of Universal Alphabetics.' In the 
preface to this work he says 

The scientific interest attaching to the invention of visible speech 
has alone induced me to consent to the publication of the system 
under copyright. My desire was that this invention the applica- 
tions of which are as universal as speech itself should at its 
inauguration have been made free from restrictions ; but my 
endeavours to effect an arrangement for this purpose have been 
frustrated. I wish to put on record here a statement of the facts 
concerning my offer of the invention to the British Government, 
and the reception of the offer. 

The author offered to relinquish all copyright in the explanatory 
work as well as all exclusive property in the system and its applica- 
tions if the expense of casting the new types and publishing the 
theory of the system were defrayed from public resources. ' This 
request was made in vain. The subject did not lie within the pro- 
vince of any of the existing State Departments, and the memorial 
was, on this ground, politely bowed out from one after the other of 
the executive offices.' 

This work was published in 1867. It was preceded in 1865 by a 
book with a similar title, ' Visible Speech, a New Fact demonstrated.' 
That the propagation of the system as a measure of public usefulness 
was still a cherished dream of the author in 1875 is evidenced by a 
letter to him from his now famous son, Alexander Graham Bell, 
who wrote on March 18 of that year recording the progress achieved 
with his harmonic telegraph. There will be occasion to refer to 
this letter again, and for the moment only the thoughtfulness of 
the son for the cherished work of the father need be noted. He 


says : ' Whenever I am free to dispose of my interest in the invention I 
shall do so, and then you may expect to see Visible Speech go ahead.' 
Both father and son were to be disappointed. The invention then 
under consideration was not destined to be directly productive, but 
the encouraging development recorded was an important aid to the 
achievement of a still greater invention. Yet though that greater 
invention has been completed, and though its use has been beyond 
the dreams of the most sanguine, Visible Speech has not ' gone 
ahead ' in the practical way which was contemplated. Its value 
in connection with the science of language may be gauged from the 
remarks made by Henry Sweet, M.A., in the preface of his work, 
' A History of English Sounds,' 1 wherein the author says that his 
investigations were due to the combined influence of Bell's ' Visible 
Speech,' Ellis's ' Early English Pronunciation,' and the German 
School of comparative and historical philology ; his use of the 
revised visible speech notation for exact purposes required, he said, 
no justification. ' Although far from perfect, it is the only system 
which is universal in its application and at the same time capable 
of being worked practically.' It was to him a source of some pride 
that, just as Henry Nicol and himself were the first to take up Bell's 
visible speech and apply it to linguistic investigation and the 
practical study of language, so also were they the first to welcome 
the revolutionary investigations in Ellis's ' Early English Pro- 
nunciation,' and he says in conclusion, ' My debt to Mr. Bell speaks 
for itself.' 

It is not, however, our province now to follow in any detail 
the indebtedness of the philologist to Alexander Melville Bell, 
but to record the important part which ' Visible Speech ' and its 
demonstrations served in the education and development of the 
author's son in the direction of acoustics and speech analysis. 
On these grounds it is necessary to give a brief account of ' Visible 

In the first place it must be said that the title has led some 
writers to assume that the subject has reference to the rendering 
visible the aerial vibrations arising from speech. The work has 
no relation to the mechanical effects of speech, nor is it a treatise 
on the instruction of deaf mutes, as another writer suggests. 2 The 
utility of the system for the deaf and dumb and also for the blind 
is subsidiary to the main purpose which, as its sub-title indicates, 
is the introduction of a universal alphabet adapted to all languages 
and all dialects. It is in fact a phonetic system as Isaac Pitman's 
phonography was a phonetic system, but it permitted the putting 
in graphic form the most varied utterances and the most delicate 

1 Oxford Clarendon Press. 1888 edition. 
1 Munro, Heroes of the Telegraph, p. 185. 


nuances of speech. Instead of adopting arbitrary characters as 
Pitman did, for an excellent reason, in his phonography, or of 
adapting existing types as in phonetic spelling, Melville Bell based 
his alphabet or signs upon a system which analysed the method 
of speech production, as for instance a circle indicating ' The Throat 
Open (aspirate),' and an oval indicating ' The Throat Contracted 
(whisper).' Other characters were based upon the position of the 
lips, the tongue, and so on, the principal object being to produce a 
written character which should be so definitely phonetic as to 
represent any variety of speech sound and thus be available for 
international usage. 

The author enumerates ten special uses of his invention, relating 
mainly to the teaching or recording of language ; but there are 
two which require to be quoted because of their bearing on the 
subsequent work of Alexander Graham Bell : 

3. The teaching of the Deaf and Dumb to speak. In this 
department very striking results may be confidently anticipated. 
The Deaf and Dumb possess all the organs of speech and only require 
to be directed visibly in their use. The feeling of organic action 
will probably be developed by practice to a keenness corresponding 
to that which the sense of touch acquires among the Blind. 

7. The Telegraphic communication of messages in any language 
through all countries, without translation. Visible Speech does not 
interfere with the use of ordinary alphabets in literature, etc., but 
for international purposes it may very advantageously supplant 
all local alphabets. Roman letters have been fully tried and found 
sadly wanting in Telegraphy. 1 

The effect of ' Visible Speech ' on the telegraphic art has been 
important but indirect rather than direct. 

Having designed his system, Mr. Melville Bell was eager to 
demonstrate its efficacy, and for this purpose taught its principles 
to his two sons, Edward Charles and Alexander Graham. The 
former died in his nineteenth year (May 17, 1867) and to his memory 
the work is dedicated. The latter happily still lives, and in the 
meantime has achieved much. Amongst the persons before whom 
the demonstrations were made was Mr. Alexander J. Ellis, F.R.S., 
who wrote an account of them for the Reader of September 3, 
1864. He recounts how the two sons who were to read the writing 
were sent out of the room whilst he dictated slowly and distinctly 
to Mr. Bell the sounds which he wished to be written. 

The result was perfectly satisfactory that is, Mr. Bell wrote 
down my queer and purposely exaggerated pronunciations and 

1 Visible Speech, pp. 20, 21. 


mispronunciations and delicate distinctions in such a manner that 
his sons, not having heard them, so uttered them as to surprise me 
by the extremely correct echo of my own voice. 

Mr. Ellis wrote a second letter to the same publication (August 5, 
1865) indicating his very great interest in, and his expectations of 
valuable results from the system. The demonstrations have had 
no lasting results on their particular subject. But the acquaintance 
of Alexander Graham Bell with Alexander Ellis was of especial 
importance, for Ellis was not only a high authority on the phonetics 
of speech, but was also the translator of Helmholtz's book, ' On the 
Sensations of Tone,' the work that first disclosed in their entirety 
the acoustic principles which underlie the production, the aerial 
transmission, and the audibility of speech. One of the direct 
results of that acquaintanceship may be seen from the statement 
made by Alexander Graham Bell that it was not until after his 
interview with Mr. Ellis in London that he really took up the 
study of the subject of electricity. 1 

Another acquaintance of Bell's early days was Sir Charles 

1 Deposition, p. 207. 



ALEXANDER GRAHAM BELL relates that Sir Charles Wheatstone lent 
to his (Bell's) father the work on ' Le Mecanisme de la Parole,' by 
Baron de Kempelen, giving a full description and plates of his 
celebrated automaton speaking machine, ' and Sir Charles Wheat- 
stone himself showed me before I came to this country [the United 
States] a reproduction of Baron de Kempelen's speaking machine 
which he had made, and I saw it operated by his own hands and 
heard it speak.' 1 

Thus was Bell early brought into communication with men 
who were most highly endowed with information on the subject 
which interested him. The son and grandson of professors of 
elocution he was himself destined for the same profession. His 
father had branched out beyond the ordinary boundaries of the 
teacher's art, had extended his interests into adjoining fields and 
saw the advantage which must come from scientific knowledge. 
Alexander Graham Bell was educated at the Royal High School of 
the Scottish capital, where he passed through the whole curriculum 
of the school. He attended lectures on classical subjects at the 
Edinburgh University, and a course on anatomy at University 
College, London, where he matriculated as an undergraduate of the 
London University in the year 1867. From his boyhood he had 
been specially educated by his father, at home, on subjects relating 
to sound and the mechanism of speech, as his father intended him 
to follow his own profession and become a teacher of articulation. 

He also received at a very early age training in music from 
Signor Auguste Bertini, after whose death his musical education 
was carried on, on the Bertini method, by his mother. 2 

From his earliest childhood his attention was directed to the 
study of acoustics, and especially to the subject of speech, and he 
was urged by his father to study everything relating to these subjects, 
as they would have an important bearing upon what was to be his 

1 Deposition, p. 207. l Ibid. pp. 6-7. 

19 C 2 


professional work. His father also encouraged him to experiment, 
and offered a prize for the successful construction of a speaking 
machine. Alexander Graham Bell made a machine of this kind, as 
a boy, and was able to make it articulate a few words. 1 He read 
all the books that his father had in his library upon these subjects, 
including his father's and grandfather's works upon speech, also 
' The Real Character,' by Bishop Wilkins, published about 1680 ; 
Prosodia Rationalis, by Joshua Steele, and the work of de Kempelen 
already referred to. 2 He recognised that he had exceptional advan- 
tages for knowing of works on speech, and that he was probably 
acquainted with a considerable portion of the literature upon that 
subject. His studies in acoustics had gone so far in 1864 or 1865, 
when teacher of elocution and music in Weston House Academy, 
Elgin, Morayshire, that he made experiments upon his own mouth 
to determine the resonance pitches of the mouth cavities during 
the production of the vowel sounds. He describes these experi- 
ments as follows : 

Considering the instrument of speech as a tube, extending from 
the vocal cords to the lips, I saw that the constriction of the 
mouth passage anywhere divided that tube into two bottle-shaped 
cavities, placed neck to neck. I came to the conclusion that these 
cavities, like ordinary bottles, should be capable of producing, 
by resonance, musical tones, and that the pitch of the tone produced 
by each cavity should be dependent upon its size, shape, etc. I 
placed the side of a lead-pencil against my cheek and tapped it 
forcibly, while I assumed with my mouth the positions for the 
vowel sounds. The agitation of the air in the front cavity of the 
mouth produced a hollow sound, like that occasioned by tapping 
against the side of an empty bottle ; and this sound had the element 
of pitch. I found that the pitch of the front cavity of the mouth 
was different for e TT ery vowel. Upon then placing the pencil against 
my throat and tapping it forcibly, as before, I caused an agitation 
of the air in the back cavity of the mouth ; and in this way was 
able to study the resonance pitches of the back cavity during the 
assumption of the vowel positions. I thus discovered that most 
vowel positions were capable of yielding a double resonance, the 
front cavity of the mouth yielding a different tone from that 
produced by the back cavity. Having familiarised myself with 
the proper tones of the mouth cavities, I experimented upon myself 
to asceitain whether I could perceive these tones while I actually 
uttered vowel sounds. Upon singing all the vowels upon the same 
pitch of voice, I was delighted to find that I could perceive the 
characteristic tones of the mouth cavities as feeble musical effects, 
mingling faintly with the voice ; and I made what I then deemed 
to be an original discovery that vowel quality was produced by 

1 Deposition, p. 7. J Ibid. p. 207. 


the resonance tones of the* mouth cavities mingling faintly with 
the tone of the voice. I made an elaborate series of experiments 
to determine the resonance pitches of the mouth cavities in uttering 
different vowel sounds, and I communicated the results to the late 
Mr. Alexander J. Ellis, of London, England. These experiments 
were made in Elgin, Scotland, somewhere about the year 1864 
or 1865. Mr. Ellis informed me that the experiments, which I 
had thought to be original with myself, had already been made by 
Helmholtz, and that Helmholtz had demonstrated the compound 
nature of the vowel sounds by producing them artificially by a 
synthetical process. For example, he would cause the simultaneous 
vibration of three tuning forks of different pitches one of these 
would represent the pitch of the voice and this fork he caused to 
vibrate in front of a resonator tuned to its own pitch so as to cause 
it to produce a loud musical tone. The other two forks corresponded 
in pitch to the front and back cavities of the mouth in uttering some 
vowel sound. These forks were caused to resound very faintly. 
The simultaneous vibration of the three forks produced one loud, 
fundamental sound, and the two higher partial tones. The effect 
upon the ear was as though someone sang a vowel sound. In an 
interview with Mr. Ellis, he attempted to describe to me the 
apparatus used by Helmholtz. Helmholtz kept his forks in 
vibration by means of electro-magnets and a voltaic battery ; but 
I found that I had not sufficient electrical knowledge to understand 
the arrangement used by Helmholtz. I therefore determined to 
study electricity ; for I felt it was my duty as a student of speech 
to study Helmholtz's researches and to repeat his experiments. 
T do not remember when I first commenced the study of electricity, 
but I know that during my stay in the city of Bath, in England, 
I was practically experimenting with ordinary telegraph apparatus, 
and trying in vain to cause the continuous vibration of a tuning 
fork by means of electro-magnets. I think the date of my stay in 
Bath was 1867, but am not quite sure. Helmholtz's researches 
were not published in the English language before my arrival in 
America ; and I could not, unfortunately, read his work in the 
original German ; but very shortly before my leaving Great Britain, 
in 1870, I procured a copy of the French edition of his work. 1 

Helmholtz's work was first published in German in 1863, in 
French in 1868. The English translation did not appear until 1875. 
The progress made by the last-mentioned date indicates that Bell 
must have been indebted to his knowledge of French for acquiring 
in detail that insight into vowel tones which was an essential 
foundation for his subsequent work. Helmholtz says : 

The vowels of speech are in reality tones produced by mem- 
branous tongues (the vocal chords) with a resonance chamber (the 

1 Deposition, p. 7. 

mouth) capable of altering in length, width, and pitch of resonance, 
and hence capable also of reinforcing at different times different 
partials of the compound tone to which it is applied. 1 

In a footnote he adds : 

The theory of vowel tones was first enunciated by Wheatstone in 
a criticism, unfortunately little known, on Willis's experiments. 
The latter are described in the Transactions of the Cambridge 
Philosophical Society, vol. iii. p. 231, and Poggendorff s Annalen 
der Physik, vol. xxiv. p. 397. Wheatstone's report on them is 
contained in the London and Westminster Review for October 1837. 

This paper of Wheatstone's is also published in his collected 
Papers, from which the following is quoted : 

Mr. Willis finally concludes, from his experiments, that the vowel 
quality, added to any sound, is merely the co-existence of its peculiar 
note with that sound ; this accompanying note being excited by 
the successive reflections of the original wave of the reed at the 
extremities of the added tube. 

This view of the matter naturally associates the phenomena of 
vowel sounds with those of multiple resonance, a subject first 
investigated by Professor Wheatstone. 

The phenomena of simple or unisonant resonance are so well 
known that we need only call attention to one or two of the most 
striking facts. If a vibrating body be brought near a column or 
volume of air, which would be capable of producing the same sound 
were it immediately caused to sound as an organ-pipe or otherwise, 
then the sound of the vibrating body is greatly reinforced, as when 
an harmonica glass is brought before an unisonant cavity, or when 
a tuning fork is placed at the embouchure of a flute, the apertures 
of which are stopped, so that, if blown into, the flute would sound 
the same note ; ir. the latter case the experiment is more remarkable 
as the sound of the tuning fork is scarcely itself audible. The 
same effect takes place when the cavity of the mouth is adjusted 
so as to be in unison with the tuning fork. 

We now come to the new facts of resonance. A column of air 
will not only enter into vibration when it is capable of producing 
the same sound as the vibrating body which causes the resonance, 
but also when the number of the vibrations which it is capable of 
making is any simple multiple of that of the original sounding body, 
or, in other words, if the sound to which the tube is fitted is any 
harmonic of the original sound.* 

The resonance effects are more fully dealt with by Wheatstone 
in his paper ' On the Resonances, or Reciprocated Vibrations 

1 Sensations of Tone, Hclmholtz. znd edition, p. 103. 
* Wheatstone's Scientific Papers, p. 358. 


of Columns of Air/ in the Quarterly Journal of Science, 1828, 
volume 3. 1 

It will be seen from the foregoing statement of Bell that his 
serious study of electricity was undertaken with a view to under- 
standing and repeating Helmholtz's experiments in the artificial 
production of vowel tones. The knowledge of electricity was 
another essential to the accomplishment of his subsequent work, 
but the application of his acoustical knowledge to the production 
of a telephone was not yet. The suggestion of Bourseul, the 
attempts of Reis, were so far unknown to him. It is probable that 
at this period Bell was amongst those who supposed that ' the 
transmission of speech required the production of a speaking 
machine such as he had in his youth attempted to develop and in 
a small measure succeeded in producing. A rough and weary road 
had yet to be travelled before he reached the goal of an idea. 

In August 1870 Mr. Melville Bell left England with his family 
and went to reside at Tutelo Heights, near Brantford, in Ontario, 
Canada. 2 In April 1871 Alexander Graham Bell was invited 
by the Board of Education of the City of Boston to make experi- 
ments in the city school for deaf mutes, to ascertain whether 
these children could be taught to speak by means of his father's 
system of visible speech. He remained two months conducting 
the inquiries at the above-mentioned and other institutions. 
During the following year, whilst residing in Canada, he paid occa- 
sional visits to the United States, and on October 1, 1872, began his 
permanent residence there. He opened, at 35 West Newton Street, 
Boston, a school of vocal physiology and received as pupils deaf 
mutes, teachers of the deaf and dumb, and hearing persons with 
defective speech. 3 

One of his pupils was a very young deaf child, named George 
Sanders ; another was a young lady named Miss Hubbard. The 
father of George Sanders and the father of Miss Hubbard jointly 
provided the funds for Bell's experiments in multiple telegraphy, and 
Miss Hubbard became the wife of the inventor of the telephone. 

Helmholtz's apparatus for the artificial production of vowel 
sounds seems to have had a sort of fascination for Bell. The experi- 
ment determined conclusively the compound nature of vowel sounds, 
the value of which Bell's previous studies enabled him to appreciate ; 
but he was impressed also with the means employed for transmitting 
the sounds. His knowledge of electricity was less than his knowledge 
of acoustics, and to him there was something new in employing one 
tuning fork to interrupt an electric current and utilising that 
electric current to vibrate another tuning fork. 

1 Wheatstone's Scientific Papers, p. 36. 2 Deposition, p. 9. 

3 Ibid. p. u. 


We are probably indebted as much to the interest of the versatile 
Samuel Pepys in music as in natural philosophy for his record of 
a conversation with Dr. Hooke, whose observation on the trans- 
mission of sounds through solid bodies was quoted in Chapter I. 
The relationship between the pitch of a tone and the frequency 
of the vibrations was explained by Hooke in a conversation with 
Pepys which took place on August 8, 1666, and is thus recorded : 

Discoursed with Mr. Hooke about the nature of sounds and he 
did make me understand the nature of musical sounds made by 
strings mighty prettily, and told me that having come to a certain 
number of vibrations proper to make any tone he is able to tell how 
many strokes a fly makes with her wings 
(those flies that hum in their flying), by 
the note that it answers to in music, 
during their flying. That I suppose is a 
little too much refined ; but his dis- 
course, in general, of sound, was mighty 
fine. 1 

One of the early examples of the 
electrical production of sound was that 
used for the purpose of determining the 
frequency with which an electro-magnet 
could be energised and de-energised. As 
' Mr. Hooke ' explained to Mr. Pepys 
that the wing flaps of a fly could be 
counted by the sound given forth, so 
M. Froment of Paris devised in 1846 2 an 
instrument in which an electro-magnet 
attracted a tongue or armature at every pulsation of current 
sent through the line, and a spring restored the tongue to its 
former position at every interruption of the current. The primary 
object of the device was not to create a sound, but to demonstrate 
the responsive nature of the electro-magnet and to determine by 
the sound resulting the number of times per second the magnet 
had been energised. The illustration of the apparatus in fig. i is 
not identical with that shown by Pouillet, but indicates the 
principle more clearly. 3 

It is unnecessary to quote the description because at this time 
the operation is generally known, but the following extract may 
be given for its reference to the effect of residual magnetism : 

The stop, /', is so placed as to prevent the absolute contact of 

1 Diary of Samuel Pepys, August 8, 1666. 

* Aliments de PhysiyiH I t/v limentale. Pouillet, i. 248. 

8 The Electric Telegraph, Lardner, 1855, p. 198. 

FIG. i. Froment's 


the arm of the lever with the electro-magnet, but to allow it to 
approach the latter very closely. Absolute contact is to be avoided, 
because it is found that in that case the arm adheres to the magnet 
with a certain force after the current ceases to flow, but so long as 
absolute contact is prevented, it is immediately brought back by 
the spring, s, when the current is suspended. 1 

By means of this apparatus it was shown 

that by the marvellously subtle action of the electric current, the 
motion of a pendulum is produced, by which a single second of 
time is divided into from twelve to fourteen thousand equal parts. 2 

Froment's experiment required that the armature or tongue 
should respond to any rate of vibration. Consequently it was 
pivoted as a pendulum and had no note of its own. From Froment 
to Helmholtz was a wide gap acoustically. Helmholtz elaborated 
the theory of resonance or sympathetic vibration, and his use of 
electricity was not to transmit a varied series of notes but to send 
such pulsations over the line as would operate a particular tuning 
fork, using electricity only as a medium of communication and 
relying on the principle of sympathetic vibration or resonance. 

A description of this principle can best be given in Helmholtz's 
own simple words. They were part of a popular lecture delivered 
during the winter of 1857 at Bonn, ' the native town of Beethoven, 
the mightiest among the heroes of harmony/ where no subject 
seemed to Helmholtz to be ' better adapted for a popular audience 
than music itself.' 

You will all have observed [he says] the phenomena of the 
sympathetic production of tones in musical instruments, especially 
stringed instruments. The string of a pianoforte when the damper 
is raised begins to vibrate as soon as its proper tone is produced in 
its neighbourhood with sufficient force by some other means. When 
this foreign tone ceases the tone of the string will be heard to 
continue for some little time longer. If we put little paper riders 
on the string they will be jerked off when its tone is thus produced 
in the neighbourhood. This sympathetic action of the string 
depends on the impact of the vibrating particles of air against the 
string and its sounding board. 

Each separate wave-crest (or condensation) of air which passes 
by the string is, of course, too weak to produce a sensible movement 
in it. But when a long series of wave-crests (or condensations) 
strike the string in such a manner that each succeeding one increases 
the slight tremor which resulted from the action of its predecessors, 
the effect finally becomes sensible. It is a process of exactly the 
same nature as the swinging of a heavy bell. A powerful man can 

1 The Electric Telegraph, I.ardner, 1855, p. 199. z Ibid. p. 201. 


scarcely move it sensibly by a single impulse. A boy, by pulling 
the rope at regular intervals corresponding to the time of its 
oscillations, can gradually bring it into violent motion. 

This particular reinforcement of vibration depends entirely on 
the rhythmical application of the impulse. When the bell has once 
been made to vibrate as a pendulum in a very small arc, and the 
boy always pulls the rope as it falls, and at a time that his pull 
augments the existing velocity of the bell, this velocity, increasing 
slightly at each pull, will gradually become considerable. But if 
the boy apply his power at irregular intervals, sometimes increasing 
and sometimes diminishing the motion of the bell, he will produce 
no sensible effect. 

In the same way that a mere boy is thus enabled to swing a heavy 
bell, the tremors of light and mobile air suffice to set in motion 
the heavy and solid mass of steel contained in a tuning fork, provided 
that the tone which is excited in the air is exactly in unison with 
that of the fork, because in this case also every impact of a wave 
of air against the fork increases the motions excited by the like 
previous blows. 1 

There was an element of novelty to Bell in combining acoustics 
and electricity. He was impressed with the employment of an 
agent to transfer that peculiar power of resonance to distance which 
seems infinity when compared with the possibilities of the air 
medium. In a word, he was intensely sympathetic to these sym- 
pathetic vibrations. His interest in music prompted him to consider 
the possibilities from a musical point of view. 

While reflecting upon the possibilities of the production of 
sound by electrical means, it struck me that the principle of vibrating 
a tuning fork by the intermittent attraction of an electro-magnet 
might be applied to the electrical production of music. 

I imagined to myself a series of tuning forks of different pitches 
arranged to vibrate automatically in the manner shown by Helm- 
holtz, each fork interrupting at every vibration a voltaic current ; 
and the thought occurred : ' Why should not the depression of a 
key like that of a piano direct the interrupted current from any 
one of these forks, through a telegraph wire, to a series of electro- 
magnets operating the strings of a piano or other musical instrument, 
in which case a person might play the tuning fork in one place and 
the music be audible from the electro-magnetic piano in a distant 
city ? 

The more I reflected upon this arrangement the more feasible 
did it seem to me ; indeed, I saw no reason why the depression of 
a number of keys at the tuning fork end of the circuit should not 
be followed by the audible production of a full chord from the piano 
with which it was in unison.* 

1 Popular Scientific Lectures, Helmholtz, p. So. 

* Journal of the Society of Telegraph Engineers, 1877, vi. 387. 


As Wheatstone, a musical instrument maker, first applied his 
acoustic discoveries to the transmission of music, so Bell, an amateur 
of music, had first the idea of utilising electricity for similar ends. 
In both cases also the development proceeded from the pleasurable 
art of music to the useful art of the communication of intelligence. 
His interest in electricity led Bell to study the various systems of 
telegraphy in Great Britain and in America. He was much struck 
with the simplicity of the Morse alphabet, and with the fact that it 
could be read by sound. He remarked that 

Instead of having the dots and dashes recorded upon paper, 
the operators were in the habit of observing the duration of the click 
of the instruments, and in this way were enabled to distinguish 
by ear the various signals. It struck me that in a similar manner 
the duration of a musical note might be made to represent the dot 
or dash of the telegraph code, so that a person might operate one 
of the keys of the tuning-fork piano referred to above, and the 
duration of the sound proceeding trom the corresponding string 
of the distant piano be observed by an operator stationed there. 
It seemed to me that in this way a number of distinct telegraph 
messages might be sent simultaneously from the tuning-fork 
piano to .the other end of the circuit, by operators each manipulating 
a different key of the instrument. These messages would be read 
by operators stationed at the distant piano, each receiving operator 
listening for signals of a certain definite pitch, and ignoring all 
others. In this way could be accomplished the simultaneous 
transmission of a number of telegraphic messages along a single 
wire, the number being limited only by the delicacy of the listener's 
ear. The idea of increasing the carrying power of a telegraph wire 
in this way took complete possession of my mind, and it was this 
practical end that I had in view when I commenced my researches 
in Electric Telephony. 1 

While the general idea of such an invention had originated 
earlier, Bell's systematic experiments in this direction did not 
begin until his residence in Boston in October i8j2. 2 The form 
of the apparatus constructed at that time consisted of tuning 
forks arranged substantially after the manner of Helmholtz. 3 The 
transmitting tuning fork was placed in a local circuit. 

Upon causing the wire to vibrate, the wire attached to the 
prong was alternately lifted out of the mercury and depressed into 
it again. The circuit of which the fork formed a part was thus 
made and broken at every vibration of the fork. The poles of 
the electro-magnet attracted the prongs of the tuning fork at each 

1 Journal of the Society of Telegraph Engineers, 1877, vi. 387. 

z Deposition, p. 12. 

3 Sensations of Tone, fig. 33, 1875 edition, p. 178 ; 1885 edition, p. 122. 


make of the circuit, and released them when the circuit was broken. 
The intermittent attraction of the electro-magnet thus caused the 
transmitting fork to remain in continuous vibration, emitting 
continuously its musical tone. By the depression of a telegraph 
key, the current rendered intermittent by the vibration of the 
transmitting fork was directed to a line wire which passed to a 
receiving instrument consisting of an electro-magnet between the 
poles of which appeared the prongs of a tuning fork. Every time 
the prong of the transmitting fork made contact with the mercury 
below it, the prongs of the receiving fork were attracted by the 
poles of the electro-magnet, between which they were placed ; 
and every time the prong of the transmitting fork broke contact 
with the mercury below, the prongs of the receiving fork were no 
longer attracted by the electro-magnet, but were allowed to move 
freely in the manner of a tuning fork left to itself. Thus, at every 
vibration of the transmitting fork, the prongs of the receiving fork 
were attracted by the receiving electro-magnet and released. When 
the receiving fork had normally the same pitch as the transmitting 
fork, the intermittent attraction of the electro-magnet would cause 
it to be thrown into vigorous vibration, thus producing a musical 
sound of similar pitch to that occasioned by the vibration of the 
transmitting fork. 1 

I proposed to use a number of transmitting forks of different 
pitches, and a number of receiving forks, each tuned to the pitch 
of one of the transmitting forks ; and I aimed so to arrange the 
pitches of these instruments that no transmitting fork should be able 
to cause vigorous vibration in the prongs of a receiving fork of 
different pitch from its own. Each transmitting fork was to be 
provided with a key, by the depression of which electrical impulses 
from the fork could be sent on to a line wire extending to some 
distant place where the receivers were to be located. The depression 
of any key would thus cause on the line wire intermittent impulses 
of electricity, which would succeed one another with the frequency 
of a musical sound ; with the frequency, in fact, of the vibration 
of the transmitting fork connected with that key. This inter- 
mittent current, passing through all of the coils of all of the electro- 
magnets of all of the receiving instruments, would occasion the 
forcible vibration of that fork alone which was the unison of the 
transmitting fork employed. That is, the receiving fork having 
the pitch of the transmitter employed would continuously emit its 
fundamental tone all the time the key at the other end was 
depressed, but the moment the key at the transmitting end was 
raised, the receiving fork would stop. Thus, if the key connected 
to any of the transmitters should be operated in the manner of an 
ordinary Morse key, long and short musical signals, corresponding 
to dots and dashes, would be emitted by that receiving fork which 
had the same pitch as the transmitter employed, but the other 
receivers would not be thrown into vigorous vibration. Thus, 

1 Deposition, p. 13. 


the manipulation of any key would cause the vigorous vibration 
of one, and only one, of the receiving instruments, so that a Morse 
message transmitted by the manipulation of one of the keys would 
be received by one, and one only, of the receivers. If two or more 
of the keys should be depressed simultaneously, then the two or 
more receiving forks corresponding in pitch to the transmitting 
forks employed would be set into vigorous vibration ; each fork 
responding to the signal sent from one, and one only, of the trans- 
mitters. Thus, upon my method, two or more telegraphic messages 
could be sent simultaneously along a single line wire, and received 
separately upon distinct receiving instruments. 1 

After a year of experimenting the tuning forks were superseded 
(November 1873) by vibrating armatures consisting of single flat 
plates really musical reeds. It was apparatus of this type which 
was chosen by Beh 1 as an illustration of his experiments in the course 
of his lecture before the Society of Telegraph Engineers in London 
on October 31, 1877. In the transmitting instrument a steel reed 
was employed which was kept in continuous vibration by the 
action of an electro-magnet and local battery. In the course of 
its vibration the reed struck alternately against two fixed points 
and so completed alternately a local and a main circuit. When 
the key was depressed an intermittent current from the main battery 
was directed to the line wire, and passed through the electro- 
magnet of a receiving instrument at the distant end of the circuit, 
and thence to the ground. The steel reed was placed in front 
of the receiving magnet, and when its normal rate of vibration 
was the same as the reed of the transmitting instrument it 
was thrown into powerful vibration, emitting a musical tone of a 
similar pitch to that produced by the reed of the transmitting 
instrument, but if it were normally of a different pitch it remained 

Subsequent illustrations in the same lecture showed the arrange- 
ment of such instruments upon a telegraphic circuit, and after 
describing them Bell proceeds : 

Without going into details, I shall merely say that the great defects 
of this plan of multiple telegraphy were found to consist, firstly, in 
the fact that the receiving operators were required to possess a good 
musical ear in order to discriminate the signals ; and secondly, 
that the signals could only pass in one direction along the line (so 
that two wires would be necessary in order to complete communica- 
tion in both directions). 2 

Here were two difficulties to be overcome. The first was accom- 
plished by a device which was termed a ' vibratory circuit breaker ' 

1 Deposition, p. 14. 

2 Journal of the Society of Telegraph Engineers, vi. 394. 


permitting the signals produced by the vibration of the musical 
reed to be reproduced upon an ordinary telegraphic instrument ; 
the second by passing the intermittent current from the trans- 
mitting instruments through the primary wires of an induction 
coil, and placing the receiving instruments in circuit with the 
secondary wire. Bell explained that 

In this way free earth communication is secured at either end of 
the circuit, and the musical signals produced by the manipulation of 
any key are received at all the stations upon the line. The great 
objection to this plan is the extreme complication of the parts, 
and the necessity of employing local and main batteries at every 
station. It was also found by practical experiment that it was 
difficult, if not impossible, upon either of the plans here shown 
to transmit simultaneously the number of musical tones that theory 
showed to be feasible. Mature consideration revealed the fact that 
this difficulty lay in the nature of the electrical current employed, 
and was finally obviated by the invention of the undulatory current. 1 

'* Journal of the Society of Telegraph Engineers, vi. 396. 



THE success which eventually attended Bell where others had 
failed may be traced to his inherited interest and special training, 
together with the thoroughness with which he investigated the 
various problems which arose and the completeness of the mastery 
which he sought over the principles of operation. The conception 
of the undulatory current was a necessity of telephonic transmission, 
but, as will have been seen from the extract at the end of the 
preceding chapter, an undulatory current was first conceived as a 
means of overcoming one of the difficulties which confronted him 
in his harmonic telegraph. Having substituted a reed for the 
tuning fork on the receiving instrument, the reed was next 
magnetised so that its vibrations should be produced by alternate 
attractions and repulsions. It was to overcome a difficulty ex- 
perienced in the transmitter which interrupted a battery current 
that Bell conceived the idea of turning the receiving reed into the 
transmitter. He says : 

It then occurred to me that the permanently magnetised reed of 
a receiving instrument should, if caused to vibrate f by mechanical 
means, itself occasion electrical impulses in the coils of its electro- 
magnet of the kind required. For I was aware of the fact that 
when a permanent magnet is moved towards the pole of an electro- 
magnet, a current of electricity appears in the coil of the electro- 
magnet ; and that when the permanent magnet is moved from the 
electro-magnet, a current of opposite kind is induced in the coils. 

I had no doubt, therefore, that a permanent magnet, like the 
reed of one of my receiving instruments, vibrating with the fre- 
quency of a musical sound in front of the poles of an electro-magnet, 
should induce in the coils of the latter alternately positive and 
negative impulses corresponding in frequency to the vibrations of 
the reed, and that these reversed impulses would come at equal 
distances apart. 1 

1 Deposition, p. 21. 


The method of operation contemplated is illustrated in fig. 2. 

M and M' are permanent magnets, E and E' electro-magnets, 
A and A' vibrating reeds of similar pitch, B and B' also vibrating 
reeds of similar pitch but differing from A and A'. The reeds are 
attached to bases in the manner usual with reed organs. The 
instruments are arranged to operate either as transmitters or 
receivers. ' In this arrangement,' Bell says, ' voltaic batteries, 
current interrupters, and induction coils became unnecessary, and 
I was fascinated with the simplicity of the arrangement of circuit.' 1 
But the generation of the electric current by such means was 
novel, and it was yet too soon for Bell to be bold enough to rely 
upon its use. 

It seemed extremely doubtful whether a magneto-electric current 
generated by the vibration of a magnetised reed in front of an 

EIG. 2. Harmonic Telegraph. (Summer of 1874.) 

electro-magnet would be sufficiently powerful to produce at the 
receiving end of the circuit a vibration sufficiently intense to be 
utilised practically, on real lines, for the purposes of multiple 
telegraphy. The arrangement seemed to fulfil perfectly the condi- 
tion I desired, namely, a succession of alternately positive and nega- 
tive impulses which should be equally distant apart. But I was 
so impressed with the idea that the magneto currents generated in 
the way proposed would be too feeble to be utilised on actual 
working lines, that I sought for some more powerful means of 
producing reversed impulses of this kind. 1 

Here was an important stage in the development of the inven- 
tion a means of creating a current of a particular character in a 
very simple way. But it was new and in the absence of experience 
it was unreasonable to expect effective results in practice. Bell 
was impressed with the idea that the currents so generated would 

1 Deposition, p. 22. 



be too feeble to be utilised under actual working conditions, and the 
line of experiment was diverted to what then seemed a more 
effective method. 

But before that further stage was reached he had considered 
the use of a series of these vibrating reeds together with their 
accompanying electro-magnets. 

I saw that the simultaneous vibration of any number of the reeds 
would produce upon the line wire a single electrical effect, the 
resultant of the effects induced in the different coils ; and that the 
curve that should graphically express this resultant would also 
represent in a graphical manner the resultant aerial effect produced 
while the reeds considered were emitting their musical tones. This 
led to the conception of an improvement in the form of the apparatus 
considered. The thought occurred instead of an electro -magnet 
coil for each reed employed, why not have one electro -magnet for 
the whole, and let the resultant effect be induced in the coil of the 


FIG. 3. Harmonic telegraph or harp telephone. (Summer of 1874.) 

electro-magnet itself, instead of inducing separate impulses in 
distinct and separate electro-magnets, and then combining the 
whole into a resultant upon the circuit ? In this way originated 
in my mind the conception of the ' harp ' apparatus shown in 
fig. 16 of my preliminary statement in the ' Harmonic Telegraph 
Interferences.' l 

Fig. 16 of the preliminary statement is fig. 19 of the Society 
of Telegraph Engineers lecture, reproduced here in fig. 3. 

My idea of the action of the apparatus, shown in fig. [3], was 
this : Utter a sound in the neighbourhood of the harp H, and certain 
of the rods would be thrown into vibrations with different ampli- 
tudes. At the other end of the circuit the corresponding rods of the 
harp H' would vibrate with their proper relations of force, and the 

1 Deposition, p. 34. 


timbre of the sound would be reproduced. The expense of con- 
structing such an apparatus as that shown in fig. [3] deterred me 
from making the attempt, and I sought to simplify the apparatus 
before venturing to have it made. 1 

Though called a harp the instrument thus suggested might be 
more accurately compared with the comb of a musical box whose 
teeth should be set in vibration sympathetically instead of by pins 
on a barrel. A compound sound being uttered in the neighbourhood 
of these teeth they would analyse that sound, and those teeth whose 
vibrations corresponded to the fundamental tone and the re- 
spective harmonics of the compound would respond with ' their 
proper relations of force/ thus originating a current of varying 
strength which would operate an identical apparatus at the receiving 
end, with the result that the originating compound sound would 
be reproduced. It is no wonder that Bell was deterred from 
constructing such an apparatus and sought a simpler method. 
Helmholtz enunciated a theory that the ear's ability to analyse 
sounds was due to the fibres within it called ' Corti's arches,' 
of which, 

according to Waldeyer, there are about 4500 outer arch fibres in the 
human cochlea. If we deduct 300 for the simple tones which lie 
beyond musical limits, and cannot have their pitch perfectly appre- 
hended, there remain 4200 for the seven octaves of musical instru- 
ments, that is 600 for every octave, 50 for every semi-tone. 2 

Upon suoh a theory, then, Bell's ' harp ' or comb would have 
needed between four thousand and five thousand ' rods ' or teeth 
to transmit speech, though a smaller number might have been 
practicable if a pre-determined pitch could have been adopted 
and retained by a speaker. 

In introducing his description of this plan at his lecture before 
the Society of Telegraph Engineers (I.E.E.) in London, Bell said : 
' I therefore devised the apparatus shown in fig. [3], which was my 
first form of articulating telephone ' ; whilst a few lines farther on 
he used the words already quoted : ' The expense of constructing 
such an apparatus as that shown in fig. [3] deterred me from making 
the attempt, and I sought to simplify the apparatus before venturing 
to have it made.' 

In this lecture Bell but lightly sketched his preliminary work, 
and the use of the word ' devised ' is probably responsible for this 
apparatus having been described as the first telephone. For 
rxainple : 

1 Journal of the Society of Telegraph Engineers, vi. 403. 
1 Sensations of Tone. p. 147. 


The first form of this instrument constructed by Professor Graham 
Bell, in 1876, is shown in fig. [3]. A harp of steel rods was attached 
to the poles of a permanent magnet. 1 

Bell's first instrument was too far from any form which has come 
into use to merit more than the briefest description here. Trans- 
mitter and receiver were similar and of harplike character, 
depending for their action upon the principle of resonance. 2 

The harp arrangement was never constructed. Bell so stated in 
his lecture and more specifically in his legal evidence. He was 
asked, ' Did you ever actually construct a harp apparatus, such 
as you have described in connection with fig. 16 of your pre- 
liminary statement in the ' Harmonic Telegraph Interferences, ' and if 
not, what mechanical difficulties were obvious ? ' To this question 
Bell replied : ' I did not, because it seemed to me impracticable, 
as a matter of fact, to construct an apparatus with the multitudinous 
reeds that theory required.' 3 

The real status of the harp apparatus is not that of the first 
telephone but the first form which occurred to Bell's mind. He says 
in his evidence : ' In this way I realised, in the summer of 1874, 
the conception of a speaking telephone, and the apparatus shown 
in fig. 16 [the harp form, fig. 3] is the first form of speaking 
telephone that occurred to my mind.' 4 

It should be added that this conception is not represented by 
the number of reeds shown in the illustration, as he also says : 

I further saw that if the reeds were multitudinous in number, 
with very slight differences of pitch between adjoining reeds, the 
reproduction of quality would be exact, and not simply approximate, 
so that whatever sound should be made or uttered in the neighbour- 
hood of harp H would be echoed in facsimile from harp H', and if 
we spoke words to harp H, the reeds of harp H' would utter words 
and reproduce the articulation of the speaker. 5 

The harp arrangement assumed the analysis of a complex sound 
into its components by the appropriate individual reeds and the 
creation by those reeds of an electric current which should be the 
equivalent in wave form of the originating sound. The apparatus 
proposed had been simplified to the extent that only one electro- 
magnet was to be employed instead of several, but the reeds which 
were to transform sound into current were still numerous. The 
quotations already given suffice to show that Bell fully realised 
that the current sent to line was a resultant just as the air 
wave was a resultant. The analysis by numerous reeds was an 

1 The Telephone, Preece & Maier, p. 21. 2 The Telephone, Hopkins, p. u. 
3 Deposition, p. 37. * Ibid. p. 37. 5 Ibid. p. 37. 

D 2 


unnecessary operation if a single reed could be so vibrated by the 
air as to produce a current which should be graphically the 
equivalent of the air wave. Reflecting on these lines it seemed to 
Bell that 

All that appeared to be necessary was, that one of the reeds, 
instead of being arranged as a free tuned reed having a rate of 
vibration of its own, should be forced to move as the air moved, 
during the production of a sound. 1 

Experiments were then in progress with a phonautograph in 
which the membrane of a human ear was used. Struck with the 
disproportion in size and weight between the membrane and the 
bones that were moved by it, Bell asked himself, ' Why should not 
a larger and stouter membrane be able to move a piece of steel in 
the manner I desired ? At once the conception of a membrane 
speaking telephone became complete in my mind ; for I saw that 
a similar instrument to that used as a transmitter could also be 
employed as a receiver.' 2 

The form which this conception assumed was that indicated 
in fig. 4, but at this period (the summer of 1874) the membrane 
form, like the harp form, was only a conception and not a 
construction. After describing the harp idea in his preliminary 
statement in the ' Harmonic Telegraph Interferences,' Bell 
says : 

Fearing that ridicule would be attached to the idea of transmitting 
vocal sounds telephonically, especially by those who were unac 
quainted with Helmholtz's experiments, I said little or nothing of 
this plan. Indeed, reflection convinced me that, however feasible 
the scheme looked upon paper, it was impracticable, as the 
induced currents would be far too feeble to overcome any great 
resistance. 3 

Bell was confident regarding the acoustic effect, but the electrical 
transmission, whether from the vibrations of one reed or from a 
series selected out of a multitude of reeds, he had no confidence in. 
The current must be too feeble to be effective, he thought, and there 
was no experience to justify any other conclusion. 

In the development of great inventions accident has often 
played a most important part. In the case of the telephone there 
is singularly little to be attributed to accidental discovery, or 

1 Deposition, p. 38. * Ibid. p. 39. 

* The Multiple Telegraph invented by A. Graham Bell, p. 13. Boston : 
Rand, Avery & Co. 1876. 


unexpected results from experiments. Bell seems to have started 
out with knowledge and devised his instruments with an accurate 
mental conception of their operation. Knowledge grew with 
experience and one step led to another, but each was thought out 
beforehand. Nevertheless, fortuitous accident helped Bell to 
revert to the right line by showing that the electro-magnetic current 
generated as by a reed was capable of more effective use than he 
had supposed. It was on June 2, 1875, that Bell and his assistant 
Watson were engaged in testing receiving instruments such as those 
described on page 32. The tests were preparatory to an experiment 
having for its object the sending of three telegraphic messages 
simultaneously along a single circuit. Three stations (A, B, and C) 
were arranged, A and B being placed ' in a small room in the upper 
part of Mr. Williams 's establishment,' * and C just outside in a room 
on the same floor. Bell was in charge of stations A and B, and 
Watson ' was observing the receiving instruments at station C.' 
At station A were three circuit-breaking transmitters, T, T 1 , T 2 , tuned 
to different pitches, and three telegraphic keys, K, K 1 , K 2 , for con- 
necting the transmitters with the line wire as desired. At station 
B were three tuned-reed receivers, R, R 1 , R 2 , having the same pitches 
as the corresponding transmitters at station A. At station C 
were also three similar tuned-reed receivers, R, R 1 , R 2 . 

Our mode of experimenting [says Bell] was as follows : I would 
depress one of the keys at station A say, key K and would 
look at the receivers at station B to see whether the corresponding 
receiver R was thrown into good vibration. At the same time Mr. 
Watson, at station C, would observe whether his receiver R was 
vibrating well. If the vibrations were not satisfactory, we would 
tune the proper receivers by shortening or lengthening the free 
part of the reed armature, according as the pitch was lower or higher 
than that desired. Sometimes we found that the reed armature 
was so closely adjusted to the pole of the electro-magnet below it 
that it would stick to the pole, instead of vibrating, the moment a 
current was passed through the electro-magnet. We would then 
cause the armature to be released by plucking it with the finger, 
and if it still stuck every time a current was passed through the coil 
of the electro-magnet, we would bend the reed armature so as to 
cause its free end to be normally a little farther away from the pole 
of the electro-magnet. On this second day of June 1875 we were 
engaged in the process of testing the receiving instruments at 
stations B and C. I depressed one of the keys at station A say, 
key K and observed the corresponding receiver R at station B. It 
seemed to vibrate well, but Mr. Watson called out that the reed of the 
corresponding receiver R at station C was sticking against the pole 
of its electro-magnet. I then told him to pluck it to release it, 

1 Deposition, p. 58. 


and he did so. At that moment I happened to have my eye upon 
the corresponding receiver R at station B, and was surprised to see 
the reed of that instrument thrown momentarily into powerful 
vibration at the very time I supposed Mr. Watson to be plucking 
the reed of his instrument. I presume that the key K of the 
corresponding transmitter T at station A was raised, for I did not 
expect the receiver R to be thrown into vibration at that time. 
Under those circumstances, the vibration of the reed immediately 
riveted my attention, and I called out to Mr. Watson to pluck his 
reed again. He did so, and again the reed of receiver R at station B 
was momentarily thrown into vibration. I kept Mr. Watson 
plucking the reed at his station, C, while I placed the reeds of the 
receivers at station B successively against my ear. At every pluck 
I could hear a musical tone of similar pitch to that produced by 
the instrument in Mr. Watson's hands, and could even recognise the 
peculiar quality or timbre of the pluck. For a long time that day 
there was little done but plucking reeds and observing the effect. 
Changes were made in the arrangements of circuit, and Mr. Watson 
and I frequently changed places. To make perfectly sure that the 
vibration of the one reed had really been the cause of the vibration 
of the other, and that the effect had been electrically produced, 
two receivers having the same pitch were arranged as in fig 5 of 
my patent l of March 7, 1876, but without any battery upon the 
circuit. Upon plucking one of the reeds with the finger, the other 
was thrown into powerful vibration vibration so strong as not 
simply to produce a musical tone, but also to be capable of oper- 
ating the vibratory circuit-breaker of my patent of April 6, 1875, 
No. 161,739, and therefore capable of use in my system of multiple 
telegraphy. These experiments at once removed the doubt that 
had been in my mind since the summer of 1874, that magneto- 
electric currents generated by the vibration of an armature in front 
of an electro-magnet would be too feeble to produce audible effects 
that could be practically utilised for the purpose of multiple tele- 
graphy and of speech transmission. 2 

It will be observed that the experiment did not reveal a new 
power but removed the doubts regarding the practical effects of 
that power. The accidental nature of the revelation is definitely 
recorded by Bell in the following letter to Mr. Hubbard : 

Salem, Mass., June 2, 1875. 

DEAR MR. HUBBARD, -I^have accidentally made a discovery 
of the very greatest importance in regard to the Transmitting 
Instruments. Indeed so important does it seem to me that I have 
written to the Organ Factory to delay the completion of the Kml 
arrangement until I have had the opportunity of consulting you. 

I have succeeded to-day in transmitting signals without any 
battery whatever ! 

1 Reproduced iii fig. 15. * Deposition, p. 58. 


The musical note produced at the Receiving End was sensibly 
the equivalent of that at the Transmitting end in loudness as well 
as in pitch. 

1 shall call upon you to-morrow (Wednesday) evening as there 
are several matters I wish to talk over with you. 

In haste, Yours respectfully, 

Brattle Street, Cambridge. 1 

This letter to Mr. Hubbard shows a great deal of self-restraint 
on the part of Bell. The importance of the discovery was fully 
realised. He was so much impressed with it that ' for a long time 
that day there was little done but plucking reeds and observing 
the effect.' The experiment was being undertaken for the purpose 
of developing multiple telegraphy, and for this it had its value ; 
but to Bell's trained intelligence it meant a great deal more. A 
transmission of sound in which timbre was produced as well as pitch 
was a practical demonstration of what he held to be possible in 
theory. The human voice he knew to be only sound with a fre- 
quently changing pitch and quality. Here was encouragement 
to press on to his goal. And encouragement was needed, for from 
his associates he received nothing but discouragement in any efforts 
towards the transmission of speech. Telegraphy was a marketable 
commodity, the transmission of speech an idle dream to his 
financial associates. The environment is evidently responsible 
for that self-restraint which permitted Bell to write of his accidental 
discovery without a word about the telephone, when his mind must 
have been full of the idea that he had reached a most important 
stage in the development of that invention. Only a week before 
(May 24, 1875) he had written to his parents : ' Every moment of my 
time is devoted to study of electricity and to experiments. The 
subj ect broadens. I think that the transmission of the human voice is 
much more nearly at hand than I had supposed.' The experiment 
of June 2 added to his confidence and encouraged his further 
energies. Six weeks later (August 14, 1875) he writes to Mr. 
Hubbard : 

On glancing back over the line of electrical experiments, I recog- 
nise that the discovery of the magneto-electric current generated by 
the vibration of the armature of an electro-magnet in front of one 
of the poles is the most important point yet reached. I believe 
that it is the key to still greater things. . . . When we can create 
a pulsatory action of the current which is the exact equivalent of 
the aerial impulses, we shall certainly obtain exactly similar results. 

1 Deposition, p. 56. 


Any number of sounds can travel through the same air without 
confusion, and any number should pass along the same wire. 

It should even be possible for a number of spoken messages to 
traverse the same circuit simultaneously, for an attentive ear can 
distinguish one voice from another, although a number are speaking 
together. 1 

The last quoted paragraph is another example of Bell's 
accurate perception and of the thoroughness with which he attacked 
his problem. The wire as the medium of transmission must carry 
the same kind or form of electrical vibrations as the aerial vibrations 
of the originating sound. These are undulatory. Hence the electrical 
current must be undulatory. He had not yet used that term, but the 
paragraph shows that he meant it, and the descriptive name was 
to follow. The stage reached at this date was indeed important. 
The undulatory current was not only realised as necessary, but 
experiment had demonstrated that it could be created, transmitted, 
and its effects reproduced. The creation was mechanical the 
plucking of the reed; the rest was electrical. There is a wide 
difference between the plucking of a reed and the creation of elec- 
trical currents by the voice. There was much ground yet to cover. 
Confidence was needed, and that confidence could only be obtained 
by Bell's own knowledge of the underlying principles and his belief 
that suitable instrumentalities could be devised. By study and 
experiment he had made much progress in the last few months in 
the electrical part of his work, encouraged thereto by the interest 
and advice of Professor Henry. 

The interview between the aged professor and the youthful 
inventor is described by Bell in a letter to his parents dated 
March 18, 1875 : 

We appointed noon next day for the experiment. I set the 
instrument working, and he sat at a table for a long time with the 
empty coil of wire against his ear listening to the sound. I felt 
so much encouraged by his interest that I determined to ask his 
advice about the apparatus I have designed for the transmission 
of the human voice by telegraph. I explained the idea, and said, 
' What would you advise me to do, publish it and let others work 
it out, or attempt to solve the problem myself ? ' He said he thought 
it was ' the germ of a great invention,' and advised me to work at 
it myself instead of publishing. I said that I recognised the fact 
that there were mechanical difficulties in the way that rendered the 
plan impracticable at the present time. I added that I felt that 
I had not the electrical knowledge necessary to overcome the 
difficulties. His laconic answer was, ' GET IT.' 

1 Deposition, p. 73. 


I cannot tell you how much^these two woids have encouraged me. 
I live too much in an atmosphere of discouragement for scientific 
pursuits. Good is unfortunately one of the cui bono people, and 
is too much in the habit of looking on the dark side of things. 
Such a chimerical idea as telegraphing vocal sounds would indeed 
to most minds seem scarcely feasible enough to spend time in working 
over. I believe, however, that it is feasible, and that I have got 
the cue to the solution of the problem. 1 

1 Deposition, p. 48. 



THE experiments of June 2, having demonstrated the efficacy of 
very weak currents, became the starting-point for further practical 
trials on more denned lines. Bell says : 

The discovery that the vibration of a steel reed in front of the 
pole of an electro-magnet generated magneto-electric currents of 
sufficient power to produce audible effects from a receiver in the 


FIG. 4. Seventh figure of Bell's patent, No. 174,465. 

same circuit, convinced me in a moment that the membrane-speaking 
telephone I had designed in the summer of 1874 would prove a 
practical working instrument. Before the second day of June 1875, 
I believed that the instrument was a theoretically perfect speaking 
telephone, but I had the idea that the currents generated by the 
action of the voice would be too weak to produce distinctly audible 
effects from the receiving instrument on a real line. The experi- 
ments made on June 2 convinced me that this idea was a mistake, 
and I immediately gave instructions to Mr. Watson to have two 
membrane telephones constructed substantially similar to those 
shown in fig. 7 of my patent of March 7, 1876 [reproduced above 
as fig. 4]. 1 

1 Deposition, p. 60. 



The apparatus here illustrated is thus described in the patent 
specification : 

The armature c, fig. [4], is fastened loosely by one extremity 
to the uncovered leg d of the electro-magnet b, and its other 
extremity is attached to the centre of a stretched membrane, a. 
A cone, A, is used to converge sound-vibrations upon the membrane. 
When a sound is uttered in the cone the membrane a is set in vibra- 
tion, the armature c is forced to partake of the motion, and thus 
electrical undulations are created upon the circuit E, b, e, f, g. These 
undulations are similar in sound to the air vibrations caused by 
the sound that is, they are represented graphically by similar 
curves. The undulatory current passing through the electro-magnet 
/ influences its armature h to copy the motion of the armature c. 

FIGS. 5 AND 6. Telephones experimented with in July 1875. 

A similar sound to that uttered into A is then heard to proceed 
from L. 1 

The first instruments that were made after the discovery of June 2 
were constructed with armatures too heavy and membranes too light. 
They gave a great deal of trouble, and instructions were given to 
have the instrument remodelled with a lighter armature and a 
stouter membrane. Shortly after July i, 1875, trial was made 
with the instruments illustrated in figs. 5 and 6, where M is a 
membrane closing tube T, and capable of being stretched drumhead 
fashion by the screws S, of which there were three, though only one 
appears in the illustration. Attached to the centre of the 
membrane is the armature A. Experiments were made with these 

1 U.S. specification. No. 174,465. 


instruments connected together in metallic circuit. Bell ' spoke 
and shouted and sang ' into one instrument upstairs while Watson 
listened to the other downstairs. Watson rushed upstairs in great 
excitement to say that he could hear Bell's voice quite plainly 
and could almost make out what was said. Bell was less successful 
as an auditor. He says : 

I do not remember the details of these experiments, nor exactly 
the results obtained, excepting that speech sounds were unmistak- 
ably produced from the receiver, and were almost intelligible, 
and that Mr. Watson appeared to hear a good deal more than I 
was able to do. 1 

The experiments were made under unsatisfactory conditions, 
and the fact that any sound at all was audible convinced Bell that 
the supposed difficulty, which had been in his mind since the 
summer of 1874, that magneto-electric impulses generated by the 
action of the voice would be too feeble to produce distinctly audible 
effects, was a mistake. The demonstration was not complete, but 
he believed that the apparatus, if carefully constructed, and tried 
in a quiet place, would transmit speech intelligibly, and prove to 
be a practically operative speaking telephone. Bell knew that he 
had conceived a remarkable invention, and he fully realised its 
importance. But that importance could only be demonstrated 
in the future. Meantime he had to live, and his arrangement with 
his financial associates only provided for the expenses of experi- 
ments and for the construction of the instruments. The inventor 
had to maintain himself by his work in other directions. He had 
been devoting considerable time to his multiple telegraph, and 
proportionately neglected the exercise of the profession to which 
he had to look for his daily bread. The ' magneto-electric current 
generated by the vibration of the armature of an electro-magnet in 
front of one of the poles ' was applicable to the multiple telegraph ; 
the ' undulatory current ' was also applicable to the multiple tele- 
graph ; but these two discoveries were clearly recognised by Bell 
as the basis of an invention far transcending in importance any 
telegraphic application. The transmission of actual speech was 
Bell's goal, but his associates were practical men ready to supply 
funds for the development of a marketable commodity like an 
improved telegraph, but shy of countenancing adventures into 
unknown, and therefore presumably unprofitable, regions. 

To proceed with experiments and adaptations so as to produce 
apparatus which would effectively demonstrate that the problem 
was solved would have been the course adopted by most inventors, 

1 Deposition, p. 70. 


and would probably have been adopted by Bell himself but for the 
series of circumstances that effectually prevented it. Conscious 
that he had conceived an important invention, that his discoveries 
included all the essential elements for carrying it out, that his 
experiments sufficiently demonstrated in practice the soundness 
of the principles upon which he was working, he decided to proceed 
at once with the preparation of a specification for a patent. The 
specification was commenced in September 1875 and substantially 
completed by the middle of the following month. 1 

The delay which arose between the completion of the specifica- 
tion in October and its filing in February further illustrates the 
difficulties with which Bell had to contend. Fully conscious 
of what he had achieved, Bell was unwilling to limit his patent 
protection to the United States alone. He wished to have 
patent protection in other countries also. Neither Mr. Hubbard 
nor Mr. Sanders was prepared to embark in foreign patents. They 
limited their outlook to the United States. Bell consequently 
had to seek support from other sources to enable him to take out 
foreign patents. For this purpose an agreement was made with 
the Hon. George Brown of Toronto and his brother Mr. Gordon 
Brown, who undertook not only to finance the foreign patents, but 
also to pay the expenses of special rooms where Bell's experimental 
apparatus could be kept private, as he was at this time troubled by 
rumours which had reached him regarding visitors to Mr. Williams's 
workshop ' examining his apparatus with curious eyes.' 2 

The specification was handed to Mr. Brown on the under- 
standing that the application should not be filed in the United 
States Patent Office until telegraphic instructions could be received 
from him. He sailed for Europe about January 25, 1876. The 
American application was sworn to in Boston on January 20, and 
was sent to Washington ' ready to be filed the moment word should 
be received from the Hon. George Brown that we might go ahead 
without interfering with his interests.' 3 But no word was received 
from Mr. Brown, and it was due to Mr. Hubbard's impatience that 
the patent solicitors filed the application on February 14, 1876. 

With some slight changes this specification became the United 
States patent No. 174,465 of March 7, 1876, which for the first time 
revealed to the world the principles of speech transmission, and 
effectually protected for the inventor the method and means 

At the time the specification was placed before the Patent Office 
the invention was in its full application a mental conception 
rather than an accomplished mechanical fact. The two membrane 
telephones previously described transmitted sounds : 

1 Deposition, p. 340, f Ibid. p. 81. a Ibid. p. go. 


The voice of the speaker was clearly perceptible from the 
receiving telephone, the intonations being heard. Articulate sounds 
were also unmistakably transmitted, but it was difficult to make 
out what was said. Mr. Watson seemed to be able to hear a good 
deal better than I could, and my impression is that he stated that 
he could almost make out what was said. 1 

But the evidence available would seem to show that no really 
convincing demonstration of the transmission of audible and 
recognisable speech was made before the deposit of the specification. 
The circumstances which prevented the active prosecution of experi- 
ments and demonstrations have already been referred to. But by 
Bell such demonstrations were not needed. He had a mental 
conception of the requirements which was so complete that actual 
demonstration seemed unnecessary. His appreciation of the value 
of the invention was equally keen, so that he wished to avoid the 
delay which might arise from the development of mechanisms to 
carry out in practice the principles which he had outlined. In his 
own words : 

I was so satisfied in my own mind that I had solved the problem 
of the transmission of articulate speech, that I ventured to describe 
and claim my method and apparatus in a United States patent, 
without waiting for better results ; in full confidence that the 
problem had been solved, and that my instruments would turn out 
to be operative speaking telephones. I was more concerned about 
taking out a caveat or patent than about further experiments. I 
believed I had experimented sufficiently to entitle me to a patent. 2 

Bell's anxieties at this time must have been considerable. 
Harassed regarding his means of livelihood, worried by fears that 
his experiments might be disclosed to his rivals in invention, it is 
easy to see that he was additionally troubled by the lack of means 
to convert his ideas into machinery and the lack of sympathy on 
the part of his associates with his ultimate aims. In these trying 
times Bell stood firmly upon his own judgment, was satisfied with 
his own knowledge of what he had accomplished, and claimed the 
protection of the patent laws of his adopted country. His applica- 
tion for a patent was allowed, and after fierce fights in later years the 
highest tribunal of that country decreed that the problem of the 
transmission of articulate speech was effectually and for the first 
time solved in the method disclosed in Bell's patent of March 7, 
1876. How complete was the solution it is hoped will be shown 
more in detail in the following pages. Some scientists, patent 
lawyers, financiers, and others more immediately interested, have 

1 Deposition, p. 338. * Ibid. p. 339. 


known that Bell's patent covered all practical methods of speech 
transmission, but the general public, influenced by the names of 
subsequent inventors of improvements in details, have regarded 
Bell as the inventor of the magneto telephone, ignorant of the fact 
that the specification which covered the employment of undulatory 
currents obtained by inductive action also covered the method of 
producing such undulations ' by gradually increasing and diminishing 
the resistance of the circuit, or by gradually increasing and 
diminishing the power of the battery,' and, in fact, the electrical 
transmission and reproduction of spoken words by such undulations, 
howsoever produced. 



RECALLING the illustration and description of fig. 7 of the patent 
(fig. 4), we may see that when the instrument is used as a 
transmitter the diaphragm operates upon the magnet through the 
medium of a suspended armature, the same operation being gone 
through in reverse order when used as a receiver. Other models 
followed on the same general lines but with improvements in details, 
like those illustrated in figs. 5 and 6. One of the principal modifica- 
tions in the fig. 6 pattern is the method adopted for supporting the 
armature. Fig. 5 followed the method of fig. 4 with a metallic 
hinge ; in fig. 6 a leather hinge was substituted, suggesting what 
seems obvious now that the metallic support or suspension was not 
sufficiently flexible to respond to the movements required. Looking 
at the subject in the light of after-knowledge, it seems a little 
surprising that a method of control involving friction or retardation 
should have been adopted. The records available do not help 
materially to disclose Bell's reasons for this form, but they may 
readily be inferred. Bell had really to go through a process of educa- 
tion by trial and experiment before he could realise how small a 
power sufficed to generate an effective telephonic current. He had 
early rejected the inductive method, believing that the power would 
be insufficient. Happy accident demonstrated the contrary, but 
he probably still supposed that an armature of substantial size 
and weight was needed, and such an armature would require 
mechanical support, for his diaphragm was a delicate fabric. 
Between the diaphragm quickly responsive to sound waves and a 
heavy armature sluggish in its movements there was a contradictory 
condition which had to be overcome. It could be overcome only by 
education the education of the inventor in the realisation of the 
fact that the current required was so insignificant. And he could 
only be his own teacher. To seek any other school would be fatal. 
The surprise expressed by the leading scientists of the day when 
the accomplishment was proved suffices to show that appeals to, 

4 8 


authorities would have been fruitless. What was needed is now clear 
enough. The armature must respond to the vibrations of the 
diaphragm with greater precision than could be expected of detached 
mechanism, but to do so its dimensions must be diminutive indeed. 
Confidence in the sufficiency of feeble curfents must be gradually 
attained. The patent was granted on March 7 although, as we 
have seen, it was prepared some months earlier. On May 10, 1876, 
Bell presented his paper, ' Researches in Telephony/ to the American 
Academy of Arts and Sciences. In section 12 of this paper he 
says : 

Two single pole electro-magnets, each having a resistance of ten 
ohms, were arranged upon a circuit with a battery of five carbon 
elements. The total resistance of the circuit, exclusive of the 
battery, was about twenty-five ohms. A drumhead of gold-beater's 
skin, seven centimetres in diameter, was placed in front of each 

FIG. 7. Centennial Single Pole Telephone (perspective). 

electro-magnet, and a circular piece of clock-spring, one centimetre 
in diameter, was glued to the middle of each membrane. 1 

Detailed information of the line of reasoning or the series of experi- 
ments which led to this result is lacking, but in the instruments 
described to the American Academy in May, and exhibited at the 
Centennial in June, we see that the hinged armature had disappeared 
and that its place was taken by a small iron disc directly attached to 
the centre of the membrane diaphragm. In another instrument 
exhibited at the Centennial the diaphragm itself was metallic. 

The instruments exhibited at Philadelphia 2 (besides a special form 
of transmitter which will be reserved for later reference) were : 

A Single-pole Magneto Telephone illustrated in perspective in 
fig. 7 and section in fig. 8. 

A Double-Pole Magneto Telephone, figs. 9 and 10, and 

An Iron Box Receiver, figs, n and 12. 

The illustrations show the instruments with sufficient clearness 

1 Proceedings of the A merican A cademy of A rls and Sciences, xii. 7. 
* Deposition, p. 96. 


to make a detailed description unnecessary. A curious feature in 
these instruments is the long cones or mouth-pieces provided. 
Asked whether he made use of ' the japanned tin cones, ' Bell said : 

I tried them and satisfied myself that they would work ; but in 
my own experiments and tests I preferred to omit them, and simply 

FIG. 8. Centennial Single Pole Telephone (section). 

speak with my mouth as near the membrane of the transmitter as 
possible. The cones or mouth-pieces were more for show than 
anything else to give a finish to the apparatus. 1 

They were apparently a tribute to convention which was not to 
last long, and it is of interest to note that Bell himself did not 
regard the cones as essential or even advantageous. 

FIG. 9. Centennial Double Pole Telephone (perspective). 

Sunday, June 25, 1876, is a notable day in telephonic history. 
The Centennial was an International Exhibition, and the judges 
were of many nations also. Amongst them were the two leading 
scientists of Great Britain and the United States, Sir William 
Thomson and Professor Henry. The latter was acquainted with 
Bell's aims, having been consulted by him, as previously recorded. 
Sir William Thomson honoured in his life by being raised to the 

1 Deposition, p. 102. 


peerage under the title of Lord Kelvin, and at his death by being 
given a resting-place in Westminster Abbey by the side of his illus- 
trious predecessor, Newton whilst covering a wide field in his 

FIG. 10. Centennial Double Pole Telephone (section). 

scientific activities was a specialist in electricity and its applications. 
But he was a stranger to Bell's efforts and a surprised witness of their 

The impression produced upon Sir William Thomson was 
recorded on the same day by Professor Hunt in the following 
letter : 

Continental Hotel, Philadelphia, 
June 25, 1876. 

DEAR Mr. BELL, I am informed that you leave tonight for 
Boston, so I take this way of congratulating you on your success 

FIG. ii. Centennial Iron Box 
Receiver (perspective.) 

FIG. 12. Centennial Iron 
Box Receiver (section). 

today. I returned to my hotel with Sir William Thomson, and 
dined with him. He speaks with much enthusiasm of your achieve- 
ment. What yesterday he would have declared impossible he has 
today seen realised, and he declares it the most wonderful thing 

E 2 


he has seen in America. You speak of it as an embryo invention, 
but to him it seems already complete, and he declares that, before 
long, friends will whisper their secrets over the electric wire. 

Your undulating current he declares a great and happy con- 

All this he discussed partly With Dr. Bache, and more at length 
with Sir Redmond Barry and Sir John Hawkshaw. Sir William 
leaves here on Friday for Montreal, and will visit Boston for a day 
or two before sailing, which will be from New York, July 19. 

Thinking you would be glad to hear the judgment of one so 
eminent, I have written you this, and I am, my dear Mr. Bell, 

Always truly yours, 


P.S. Do you know anything of Briicher's system of visible 
speech, of which one of the Austrian judges spoke to-day ? It seems 
very like your father's. 

Whether by any rights of office or merely by the tacit recognition 
of their colleagues that their qualifications were supreme is not 
recorded, but Henry and Thomson were delegated to draft the 
judges' reports. It was they who realised the scientific achievement 
and the practical utility. The General Report of the Judges 
was written by Henry. After reference to the multiple telegraph 
he says : 

The telephone of Mr. Bell aims at a still more remarkable result 
that of transmitting audible speech through long telegraphic 
lines. In the improved instrument the result is produced with 
striking effect, without the employment of an electrical current 
other than that produced by the mechanical action of the impulse 
of the breath as it issues from the lungs in producing articulate 
sounds. To understand this wonderful result, suppose a plate of 
sheet iron, about five inches square, suspended vertically before 
the mouth of the speaker so as to vibrate freely by the motion of 
the air due to the speech, and suppose also another iron plate, of 
the like dimensions, similarly suspended before the ear of the hearer 
of the sounds, and between these, but not in contact with them, is 
stretched the long telegraphic wire. Each end of this wire is 
attached to two coils of insulated wire surrounding a core of soft 
iron, the ends of which are placed near the middle of the plate, but 
not in contact with it. These four cores are kept in a magnetic 
condition by being attached at each end of the line to the two poles 
of a permanent magnet. Now it is evident that in this arrangement 
any disturbance of the magnetism of one of the permanent magnets, 
increasing or diminishing it, will induce electrical currents, which, 

1 Deposition, p. 101. 


traversing the long wire, will produce a similar disturbance of the 
magnetism of the arrangement at the other end of the wire. Such 
a disturbance will be produced by the vibration of the plate of soft 
iron due to the words of the speaker, and the current thus produced, 
changing the magnetism of the soft iron cores, will by reaction 
produce corresponding vibrations in the iron plate suspended before 
the ear of the hearer. The vibrations of the second plate being 
similar to those of the first will reproduce the same sounds. Audible 
speech has, in this way, been transmitted to a distance of three 
hundred miles, perfectly intelligible to those who have become 
accustomed to the peculiarities of certain of the sounds. All parts 
of a tune are transmitted with great distinctness and with magical 

This telephone was exhibited in operation at the Centennial 
Exhibition, and was considered by the judges the greatest marvel 
hitherto achieved by the telegraph. The invention is yet in its 
infancy and is susceptible of great improvements. 1 

The Report on Awards was written by Thomson. This also 
refers to the multiple telegraph, and then proceeds : 

In addition to his electro-phonetic multiple telegraph, Mr. 
Graham Bell exhibits apparatus by which he has achieved a result 
of transcendent scientific interest the transmission of spoken 
words by electric currents through a telegraph wire. To obtain 
this result, or even to make a first step- towards it the transmission 
of different qualities of sounds, such as the vowel sounds Mr. Bell 
perceived that he must produce a variation of strength of current 
in the telegraph wire as nearly as may be in exact proportion to 
the velocity of a particle of air moved by the sound ; and he invented 
a method of doing so, a piece of iron attached to a membrane, 
and thus moved to and fro in the neighbourhood of an electro- 
magnet, which has proved perfectly successful. The battery and 
the wire of this electro-magnet are in circuit with the telegraph 
wire and the wire of another electro-magnet at the receiving station. 
This second electro-magnet has a solid bar of iron for core, which 
is connected at one end, by a thick disc of iron, to an iron tube 
surrounding the coil and bar. The free circular end of the tube 
constitutes one pole of the electro-magnet, and the adjacent free 
end of the bar-core the other. A thin circular iron disc, held pressed 
against the end of the tube by the electro-magnetic attraction, and 
free to vibrate through a very small space without touching the 
central pole, constitutes the sounder by which the electric effect is 
reconverted into sound. With my ear pressed against this disc, 
I heard it speak distinctly several sentences, first of simple mono- 
syllables, ' To be or not to be ' (marvellously distinct) ; afterwards 
sentences from a newspaper, ' S. S. Cox has arrived ' (I failed to 
hear the ' S. S. Cox/ but the ' has arrived ' I heard with perfect 

1 General Report of Judges, Centennial Exhibition, Group XXV. p. 20. 


distinctness) ; then ' City of New York,' ' Senator Morton/ ' The 
Senate has passed a resolution to print one thousand extra copies,' 
' The Americans of London have made arrangements to celebrate 
the fourth of July.' I need scarcely say I was astonished and 
delighted ; so were the others, including some judges of our group 
who witnessed the experiments and verified with their own ears 
the electric transmission of speech. This, perhaps the greatest 
marvel hitherto achieved by the electric telegraph, has been obtained 
by appliances of quite a homespun and rudimentary character. 
With somewhat more advanced plans and more powerful apparatus, 
we may confidently expect that Mr. Bell will give us the means of 
making voice and spoken words audible through the electric wire 
to an ear hundreds of miles distant. 1 

What could have been more appropriate than that the leading 
scientist of America, who at an earlier stage had given much needed 
encouragement to persevere by expressing the opinion that the 
experiments described by Bell contained ' the germ of a great inven- 
tion,' 2 and the leader of British scientists, should be the first to 
express appreciation of the development of that germ, to put the seal 
of science upon that great invention ? Popular applause, however 
great, can never convey a gratification equal to the appreciation of 
the well-informed. Alexander Graham Bell experienced this grati- 
fication in full measure. 

Two months later Sir William Thomson presided over the 
meeting of the Mathematical and Physical Section of the British 
Association for the Advancement of Science at Glasgow, and in the 
course of his presidential address narrated his experience at the 
Centennial Exhibition, repeating much of his judges' report, and 
adding : 

This, the greatest by far of all the marvels of the electric telegraph, 
is due. to a young countryman of our own, Mr. Graham Bell, of 
Edinburgh and Montreal and Boston, now becoming a naturalised 
citizen of the United States. Who can but admire the hardihood 
of invention which devised such veiy slight means to realise the 
mathematical conception that, if electricity is to convey all the 
delicacies of quality which distinguish articulate speech, the strength 
of its current must vary continuously and as nearly as may be in 
simple proportion to the velocity of a particle of air engaged in 
constituting the sound. 3 

The first impressions, briefly recorded by Professor Hunt, 
are enlarged but unaltered. One of the greatest mathematicians of 
his time was impressed with the ' mathematical conception,' the 

1 Reports on Awards, Centennial Exhibition, Group XXV. p. 130. 
1 Page 40. Nature, Sept. 14, 1876, xiv. 427. 


greatest electrician of his age recognised the perfection of the 
means for realising that conception. There was an element of 
good fortune in the fact that Sir William Thomson was one of the 
judges at the Centennial. No one had as yet dared to contemplate 
the possible field of usefulness for the invention, the financial 
associates were still pressing for multiple telegraph instruments, 
the inventor even now stood alone in his faith in the commercial 
developments of the telephone, and the words of Sir William Thomson 
were not only encouraging to him but were also helpful in their 
justification for pushing along further development. They informed 
the world on authority which permitted no carping doubts that 
speech had been transmitted, and that this result had been reached 
by the application of true scientific principles. At the achievement 
the great scientist marvelled, and the world marvelled with him. 



ON July 12, 1876, in the presence of Sir William Thomson, experi- 
ments were made between neighbouring rooms in the Equitable 
Buildings at Boston over a circuit extending to New York and 
back. The condition of the line rendered the experiment unsatis- 
factory, but on short circuiting the New York loop the articulation 
became audible. After the conclusion of the experiment Bell 
presented the instruments to Sir William Thomson as a memento 

FIG. 13. Double Pole Instrument. 
(July 1876). 

FIG. 14. Iron Box Receiver (as 
exhibited at Glasgow). 

of the occasion, and they were exhibited by him to the members 
of the British Association at Glasgow. 

These instruments followed closely the design of those exhibited 
at the Centennial but varied in some respects. The electro-magnet 
of the double pole instrument (fig. 13) was larger and longer than 
the Centennial and was wound with finer wire. The iron box 
receiver (fig. 14) was of larger diameter than the Centennial and, as 
will be noted from tHe illustration, was differently mounted upon 
its base. It had one other difference, as exhibited at Glasgow, 
which is responsible for much mistaken history. The difference 
is related by Bell : - 

The iron diaphragm or lid has a small screw at one side fastening 
it to the edge of the iron box. The Centennial iron box receiver 



had no such screw. The instrument used on July 12, 1876, had this 
feature. The screw was not employed during the course of experi- 
ments, but when the instrument was not in use, the diaphragm was 
screwed to the edge of the iron box so that it should not be lost. 
At the conclusion of the experiments on July 12, we had no time 
to pack the telephones for transportation to England. I simply 
gave them to Sir William, and he said he would put them in his 
trunk. The diaphragm of the iron box receiver, not being protected 
by a packing case, was probably injured on the voyage, so that 
upon arrival in England the iron lid was somewhat distorted, and did 
not lie down flat on the top of the iron box as it should have done. 1 

Sir William Thomson's address at Glasgow was reported in 
Engineering of September 15, 1876, and the instruments exhibited 
were illustrated in the issue of that journal for December 22, 1876. 
Sir William was evidently unaware that the screw was not intended 
to remain when the instrument was in use, or that the tilt in the 
diaphragm was accidental. He gave evidence in the case of the 
United Telephone Co. Ltd. v. Alexander Maclean tried before Lord 
M'Laren in the High Court at Edinburgh in January 1882. 
Speaking of the Philadelphia experiment he said : ' The only way 
it could be heard was by the ear being pressed upon the disc. If 
pressed too heavily it would kill the sound, and if too little 2 the 
same result ensued. The great difficulty was adjusting the ear.' 3 
From the general tenor of his evidence in this case it seems clear 
that Sir W'illiam Thomson failed to remember that the disc was 
intended to touch the tube throughout its circumference. The 
drawing in Engineering accurately represented the instrument 
as exhibited at Glasgow. The British Association meeting was its 
introduction to the scientific world, and so it happens that the 
illustrations of the Glasgow instruments became incorporated in 
history. The drawing with the tilted disc went back to the United 
States and was included in Prescott's ' Speaking Telephone,' 1878 
and 1884 ; it appears in Du Moncel's ' The Telephone, the 
Microphone, and the Phonograph,' 1879, and all other histories 
and text-books. It so appears even in the report of Bell's lecture 
of 1877 in the Journal of the Society of Telegraph Engineers, 
vol. vi. p. 408, whence the illustration (fig. 14) is taken. 

The circumstance was explained by Sir William in his evidence 
in the Telephone Case of 1882, but in telephonic literature the 
error was not publicly corrected until the third edition of Mr. 
Kempster Miller's ' American Telephone Practice/ published in 
1900, which contained a letter from Mr. T. D. Lockwood reciting 
the facts. 

1 Deposition, p. 116. z ? lightly. 

8 Electrical Review, London, Jan. 28, 1882, x. 70. 


As was stated in the Edinburgh case, these Glasgow instruments 
could not be made to work. The reason is to be found in the 
illustration. It was presumed that the diaphragm was bent 
intentionally instead of accidentally, that the screw fulfilled a 
a permanent, instead of temporary, purpose. In operation the 
screw should have been removed and the diaphragm permitted 
to rest evenly on the box throughout its circumference, as shown 
in the illustration of its predecessor (fig. 12). The tubular form 
of magnet was only reached by stages. Bell had heard articu- 
late speech produced by a tuned reed, fig. 15 (fig. 5 of patent, 
March, 7, 1876), when the reed was sufficiently damped by 
close contact with his ear, and he thought that he could reduce 
the tendency of the reed to vibrate to its own tone by clamping 
it firmly at both ends and subjecting the centre of the reed to the 
attraction of an electro-magnet. He devised a receiver having 
a magnet with three parallel cores upon a single yoke. The 
centre core was to have a coil wound upon it, and the two 

FIG. 15. Tuned Reed (fig. 5 of patent, March 7, 1876). 

outside cores were to support a bridge of steel which should 
pass over, without touching, the centre core. One great 
advantage that occurred to Bell in this connection was that the 
opposite poles of the electro-magnet would be very close together ; 
the end of the covered leg constituting one pole, and the centre of 
the steel armature the other. 1 

From this stage he proceeded to consider the advantages of a 
circular plate and an increased number of limbs to the magnet, when 
he came across a tubular magnet in Mr. Williams's shop and realised 
that a tubular form would be the best. Professor Silvanus P. 
Thompson calls this form of magnet ' the Ironclad electro-magnet,' 
and says of it : 

The appropriate armature for electro-magnets of this type is 
a circular disc or lid of iron. It is curious how often the use of a 
tubular jacket to an electro-magnet has been reinvented. It dates 
back to about 1850, and has been variously claimed for Romers- 
hausen, Guillemin, and for Fabre. It is described in Davis's 
' Magnetism,' published in Boston in 1855. About sixteen years 
ago Mr. Faulkner, of Manchester, revived it under the name of the 

1 Deposition, p. 321. 


Allandce electro-magnet. A discussion upon jacketed electro- 
magnets took place in 1876 at the Society of Telegraph Engineers ; 
and in the same year Professor Graham Bell used the same form 
of electro-magnet in the receiver of the telephone which he exhibited 
at the Centennial Exhibition. 1 

It is to be borne in mind, however, that Bell's use of this form 
was not the ordinary use of it. In normal conditions the ironclad 
magnet has its centre and circumference in the same plane, and 
the armature makes contact with the centre as well as with the 
circumference. It will be seen on reference to the sectional illus- 
tration (fig. 12, p. 51) that the central core in Bell's instru- 
ment was adjustable, and the object of the adjustment was that the 
core should approach the diaphragm as nearly as possible without 
touching it. The iron box receiver is an important starting-point 
for subsequent developments, but as illustrated in the Glasgow 
pattern an erroneous idea of its operation is conveyed. The tilt 
of the disc gives it the appearance of a beating reed instead of a 
diaphragm whose circumference is in contact with the iron tube 
and whose centre is opposed to the core. In Bell's own words : 

By placing upon it a lid or diaphragm of iron or steel, we had an 
armature that was damped all around, and polarised by contact 
with the rim of the box. The end of the central core would then 
constitute one pole of the magnet, and the centre of the diaphragm 
or lid the opposite pole. This seemed to me an especially advan- 
tageous arrangement. This was the origin of the iron box receiver, 
and in order to damp the vibrations of the lid or diaphragm still 
further, the ear was placed closely against the lid, as it had been in 
former experiments against the reed of the tuned reed receiver. 
This was the way in which the instrument was generally employed, 
and it was so used at the Centennial Exhibition. The listeners were 
instructed to press their ears firmly against the lid of the iron box 
receiver. 2 

It was customary with Bell throughout his experiments to 
use his various instruments interchangeably as transmitters and 
receivers. The iron box receiver and the Centennial transmitter 
had both been thus reversibly operated. The best results were 
obtained when the membrane telephone was used as a transmitter 
and the iron box as a receiver. 3 The reasons were that the ear could 
not be applied directly to the membrane telephone when used as a 
receiver, and that the iron box instrument had a diaphragm ' of 
such small diameter and such stiff material that it was not as well 
adapted to be vibrated by the voice as the membrane instrument.' 4 

1 The Electro-magnet, Silvanus P. Thompson, p. 52. 

* Deposition, p. 103. 3 Ibid. p. 104. * Ibid. p. 62. 


The diameter of the Centennial iron box diaphragm was 
if inch, the Glasgow pattern, fig. 14, p. 56, was larger, whilst 
the diameter of the membrane telephone diaphragm was 
3 inches. 

The greater flexibility of the membrane and the readiness with 
which its tension could be adjusted seemed to render it more 
suitable than a metallic plate. Yet the metallic diaphragm had 
been used for the receiver, and the success with which speech had 
been transmitted proved that the flexibility of the iron diaphragm 
was equal to the requirements so far as the reproduction of speech 
was concerned. It is probable that the realisation of the 
minute current which would do the work was not yet complete, 
and that the membrane remained the favourite for transmitters 
because of the greater excursions that could be expected from it. 
But the superiority of the metallic plate was gradually being 

Bell had used voltaic batteries in circuit with his instruments 
in his earlier work, but soon came to realise that the function of the 
battery was merely to energise the electro-magnet, and that a 
permanent magnet might be substituted for the latter, thus ren- 
dering the battery superfluous. In July 1876, therefore, he had an 
instrument constructed similar in pattern to the Centennial double 
pole telephone, but employing permanent magnets instead of 
electro-magnets with battery, and after a little experimenting the 
battery was discarded and the permanent magnet relied upon. The 
diaphragm, however, continued to be of a membrane form with a 
metallic patch. But repeated experimental demonstrations gave 
confidence in the metallic diaphragm, and on January 15, 1877, an 
application for a second patent was filed, which contains the following 
paragraph : 

In my patent No. 174,465, dated March 7, 1876, I have shown 
as one form of transmitting instrument a stretched membrane, to 
which the armature of an electro-magnet is attached, whereby 
motion can be imparted to the armature by the human voice, or 
by means of a musical instrument, or by sounds produced in any 
way. In accordance with my present invention I substitute 
for the membrane and armature shown in the transmitting and 
receiving instruments alluded to above, a plate of iron or steel 
capable of being thrown into vibration by sounds made in its 
neighbourhood. 1 

The same specification introduces the permanent magnet : 

1 U.S. specification, No. 186,787, January 30, 1877 (application filed 
January 15, 1877). 


instruments without a battery by rendering the central bar FH 
[fig. 16] permanently magnetic. Another form of telephone for use 
without a battery is shown in fig. [17], in which O is a compound 
permanent magnet, to the poles of which are affixed pole pieces of 
soft iron, P Q, surrounded by helices of insulated wire, R S. 1 

The eighth claim combines the two features of permanent magnet 
and metallic diaphragm in the following words : 

In a system of electric telephony, the combination of a perma- 
nent magnet with a plate of iron or steel, or other material capable 
of inductive action, with coils upon the end or ends of said magnet 
nearest the plate, substantially as set forth. 1 

FIG. 16. Box Telephone with Bar 
Magnet (1877 patent). 

FIG. 17. Double Pole 
Type (1877 patent). 

This second patent is numbered 186,787, and was granted 
January 30, 1877. 

The drawings in this specification show the continuance of a 
tendency to rely on boxes and long mouth-pieces, but when the 
combination of the permanent magnet and iron diaphragm had been 
reached, it was clearly but a matter of arrangement to dispose these 
elements in the most suitable way, and Bell devoted his attention 
to perfecting the form. 

The double purpose served by the single instrument is thus 
definitely stated : 

The sender of the message will use an instrument in every 
particular identical in construction and operation with that em- 
ployed by the receiver, so that the same instrument can be used 
alternately as a receiver and a transmitter. 1 

The Centennial Exhibition instruments, however, differed 
according to their intended use as transmitter or receiver, and this 
difference in form was continued, although the principles of 
operation were the same. 

The transmitting instrument was called the Box Telephone, and 
is illustrated in fig. 18 with the cover removed. 

1 U.S. specification, No. 186,787, Jan. 30, 1877 (app. filed Jan. 15, 1877). 


In a pamphlet 1 from which this illustration is taken the date of 
June 1877 is assigned to it, though Prescott 2 says that the instru- 
ment used in talking between Boston and Somerville in April 1877 
was essentially like it. 

FIG. 18. Box Telephone with cover removed (June 1877). 

A plan view of this instrument is given in fig. 19. 

The growing commercial use of the invention would doubtless 
soon demonstrate the somewhat inconvenient form of this box 
telephone, so that within two or three months a new model was 

FIG. 19. Box Telephone (plan view). 

evolved, the only change being a modification in the arrangement of 
the parts. 

When the diaphragm and mouth-piece were placed at the end of 
the magnet and in line with it, as in the earlier model, a long box 

1 Alexander Graham Bell, the inventor of the Electric Speaking Telephone, 
undated, but probably issued in 1881. 

* Bell's Electric Speaking Telephone, Prescott, 1884, p. 440. 


was necessary. Attached to a wall it would project unduly. The 
inconvenience of this form in commercial use was doubtless promptly 
recognised, for by August 1877 the model illustrated in fig. 20 
was brought out. By placing the pole pieces and diaphragm at right 
angles to, instead of in line with, the magnet the instrument was well 
adapted to attach to a wall, and this constituted the transmitter. 

The first hand telephone fol- 
lowed the box telephone in being 
of the double pole type. A 
U-shaped permanent magnet 
was placed inside a cylindrical 
wooden case of small enough dia- 
meter to be easily grasped by 
the hand, and thus form a handle 
by which the telephone could be 
lifted to the mouth or ear as 
required. 1 An instrument of this 
form was publicly exhibited at 
a meeting of the Society of Arts 
in Boston in May 1877, and 
though the first to be completed, 
it was one of four ordered at the 
same time, 'two to contain U- 
shaped permanent magnets, and 
two to contain straight -bar per- 
manent magnets,' and all were 
to be of the same general shape. 
There are no illustrations of the 
double pole hand instrument, but 
by May 1877 the straight bar 
model had assumed the form 
shown in fig. 21 : 

Another model (fig. 22) was 
issued in June, the change being 
apparently limited to the exterior 
of the case. 

But by this time scientific interest had been aroused, and 
numerous people were investigating causes and suggesting designs. 

Amongst others were a number of professors of Brown University 
at Providence, who conducted experiments in the Physical Labora- 
tory of that institution, and kept Bell advised of their progress. 
Of the work of these gentlemen Bell thus spoke in his London 
lecture to the Society of Telegraph Engineers : 

FIG. 20. Box Telephone 

(August 1877). 

1 Deposition, p. 155. 


And here I wish to express my indebtedness to several scientific 
friends in America for their co-operation and assistance. I would 
specially mention Professor Peirce and Professor Blake, of Brown 

FIG. 21. Hand Telephone (May 1877). 

University, Dr. Channing, Mr. Clarke, and Mr. Jones. In Pro- 
vidence, Rhode Island, these gentlemen have been carrying on 
together experiments seeking to perfect the form of apparatus 
required, and I am happy to record the fact that they communicated 
to me each new discovery as it was made, and every new step 
in their investigations. It was, of course, inevitable that these 

FIG. 22. Hand Telephone (June 1877). 

gentlemen should retrace much of the ground that had been gone 
over by me, and so it has happened that many of their discoveries 
had been anticipated by my own researches ; still, the very honour- 
able way in which they from time to time placed before me the 
results of their discoveries entitles them to my warmest thanks, and 
to my highest esteem. It was always my belief that a certain ratio 


would be found between the several parts of a telephone, and that 
the size of the instrument was immaterial ; but Professor Peirce 
was the first to demonstrate the extreme smallness of the magnets 
which might be employed. And here, in order to show the parallel 
lines in which we were working, I may mention the fact that two 
or three days after I had constructed a telephone of the portable 
form, containing the magnet inside the handle, Dr. Channing 
was kind enough to send me a pair of telephones of a similar 
pattern, which had been invented by the Providence experimenters. 
The convenient form of mouth-piece shown in fig. [22]. now 
adopted by me, was invented solely by my friend Professor 
Peirce. 1 

FIG. 23. Hand Telephone (December 1877). 

By April 1877 the telephone had ' got into the factory.' Some 
time previous to April i, 1877, Bell and his associates made an 
arrangement with Charles Williams, junior, of Boston, for the manu- 
facture of his telephones with metallic diaphragms for general 
commercial use. 

After other slight changes in shape in the meantime, the hand 
telephone assumed its more definite form (fig. 23) in December 1877, 
when wood as a material for the handle was given up and hard 
rubber adopted instead. For the magnet, compound bars made 
up of several layers of magnetised steel were used. 

This was by no means the last change to be made in the form 
of the telephone as a receiver, but we cannot now follow these 
changes further. The instrument had reached a commercial form ; 
it was to be put to commercial uses. The direction or extent of 
those uses few, if any, could foresee. 

1 Journal of the Society of Telegraph Engineers, vi. 411. 



COMMERCE is, in the main, conservative. The introduction to 
public attention of an adaptation of some existing condition is 
more promising of success than the attempt to create a new demand. 
The practical applications of the transmission of intelligence have 
followed this general rule. ' Pantomime/ said Bernardin de St. 
Pierre, ' is the first language of man,' and pantomime was the 
first means of transmitting intelligence to a distance because 
the power of the unaided eye is more far-reaching than that of the 
unaided ear. So man communicated ideas to distant man by the 
waving of arms. When Chappe was impressed with the possibility 
of improving the means of communication his conception took the 
form of replacing the arms of the individual with arms on an elevated 
post. The number of links in the chain of communication was 
reduced, and thereby economy in time and money was effected. 
A code was established which put into language the wavings of 
semaphore arms. So when Wheatstone and his co-pioneers essayed 
the transmission of intelligence electrically, they employed elec- 
tricity to move needles. The method of transmission was new, but 
the mental operations were along familiar lines. 

What more natural than that the proprietors of the telephone 
should seek to obtain public support by grafting it upon an existing 
stock rather than awaiting the development of an independent 
root ! The telegraph existed, but trained operators were necessary 
for its use. To supersede the telegraph instruments by telephones 
would not only be economical on some existing lines, but the substitu- 
tion of the spoken word for a cumbrous code would materially enlarge 
the field of utility. The name first given to the telephone by the 
public was the ' talking telegraph ' ; the first public use was expected 
to be that of a talking telegraph. 

So soon as a practical commercial instrument had been com- 
pleted, the proprietors of the patents went into business. They 



lost no time in issuing a circular, of which the following is a 
copy : 


The proprietors of the Telephone, the invention of Alexander 
Graham Bell, for which the patents have been issued by the United 
States and Great Britain, are now prepared to furnish Telephones 
for the transmission of articulate speech through instruments not 
more than twenty miles apart. Conversation can be easily carried 
on after slight practice and with the occasional repetition of a word 
or sentence. On first listening to the Telephone, though the sound 
is perfectly audible, the articulation seems to be indistinct ; but 
after a few trials the ear becomes accustomed to the peculiar sound 
and finds little difficulty in understanding the words. 

The Telephone should be set in a quiet place, where there is no 
noise which would interrupt ordinary conversation. 

The advantages of the Telephone over the Telegraph for local 
business are : 

1. That no skilled operator is required, but direct communica- 
tion may be had by speech without the intervention of a third 

2. That the communication is much more rapid, the average 
number of words transmitted a minute by Morse Sounder being 
from fifteen to twenty, by Telephone from one to two hundred. 

3. That no expense is required either for its operation main- 
tainance or repair. It needs no battery, and has no complicated 
machinery. It is unsurpassed for economy and simplicity. 

The terms for leasing two Telephones for social purposes 
connecting a dwelling house with any other building will be $20 
a year, for business purposes $40 a year, payable semiannually in 
advance, with the cost of expressage from Boston, New York, 
Cincinnati, St. Louis, or San Francisco. The instruments will be 
kept in good working order by the lessors, free of expense, except 
from injuries resulting from great carelessness. 

Several Telephones can be placed on the same line at an 
additional rental of $10 for each instrument ; but the use of more 
than two on the same line where privacy is required is not advised. 
Any person within ordinary hearing distance can hear the voice 
calling through the Telephone. If a louder call is required one can 
be furnished for $5. 

Telegraph lines will be constructed by the proprietors if desired. 
The price will vary from $100 to $150 a mile ; any good mechanic 
can construct a line ; No. 9 wire costs 8 cents a pound, 320 pounds 
to the mile ; 34 insulators at 25 cents each ; the price of poles and 
setting varies in every locality ; stringing wire $5 per miles ; sundries 
$10 per mile. 

Parties leasing the Telephone incur no expense beyond the 
annual rental and repair of the line wire. On the following page 

F 2 


are extracts from the Press and other sources relating to the 


Cambridge, Mass., May, 1877. 

For further information and orders address 

THOS. A. WATSON, 109 Court St., Boston. 1 

The business tone of the circular is commendable. There is no 
attempt at exaggeration either as regards the distance to be traversed 
or the audibility of the speech. Another feature to be noticed is 
that at this early stage instruments were to be let out on hire, not 
sold outright. 

A pictorial representation of the use of the telephone with two 
instruments on a direct line and with ' more than two where privacy 
is not required ' was given in the Scientific American of October 6, 
1877, illustrating an article entitled ' The New Bell Telephone.' 
Fig. 24 is a reproduction of this illustration. 

One hundred and nine Court Street, Boston, whence the first 
circular was issued, was also the workshop where the telephone had 
been evolved. Thomas A. Watson, to whom communications were 
to be addressed, had been the close associate of the inventor during 
its evolution. Charles Williams, an electrical jnanufacturer when 
such manufacturers were few, was the occupier of these premises. 
Here he conducted his business, and the first line constructed 
exclusively for telephone use was built between Charles Williams's 
suburban home at Somerville and his oifice in Court Street. It was 
completed April 4, 1877, and was noticed in the Boston newspapers 
of the following day. 

This line was promptly followed by another connecting Williams's 
office with Bell's laboratory at No. 5 Exeter Street, Boston. Two 
other lines were constructed from Exeter Street to the offices of 
persons with whom Bell was associated, and in the early part of 
May 1877 an arrangement was made with the Cambridge Board of 
Waterworks to put up a line connecting their office in Cambridge 
with the works at Fresh Pond. This line was constructed for the 
practical business purposes of a customer, and the use of the tele- 
phone in connection with it was referred to in the Boston Herald 
and Advertiser of May 19, 1877. 

In New York a telephone line was erected on May 18, 1877, 
between the house and office of H. L. Roosevelt ; and on May 21, 
telephones were permanently placed on a line at Altoona, 

By June 30, 1877, 230 telephones were in regular use ; this number 

1 Copied from The Electrical World and Engineer, March 5, 1904 (vol. xliii. 
p. 447), which published a reduced facsimile from an original copy in the 
possession of Mr. T. C. Martin. 



within one month had increased to upwards of 750 ; at the end of 
August to I3OO. 1 These were mainly as substitutes for telegraph 
instruments for communicating between two points. But already 
there had been a glimmering of the possibility that might await an 



=&** H 


<u -, 

interchangeable system. There existed in Boston, as elsewhere, 
a system of electric burglar alarms ; doors and windows were 
guarded by contact making-and-breaking devices, a signal being 
given when a contact was broken. Buildings so fitted were con- 
nected with a central point, and the signal was given there. 

The Holmes Burglar Alarm Co. had a central station at 342 

Boston Electrical Handbook (1904), p. 128. 


Washington Street, Boston, whence burglar alarm lines radiated to 
a number of banks and stores. Arrangements were made for the 
use of these lines, their sub-stations and the central station as an 
experimental telephone exchange. 

The lines of Brewster, Bassett & Co., bankers (now Estabrook 
& Co.), the Shoe and Leather Bank, the National Exchange Bank, 
and the Hide and Leather Bank, together with a new line from the 
office of Mr. Williams, the manufacturer, were fitted out with 
telephones and connected at the Holmes central station with a small 
switchboard made for the purpose. These lines were repeatedly 
interconnected, and many conversations were interchanged between 
their stations, the burglar alarm apparatus being employed to 
transmit the regular call signals. This was, in fact, the first telephone 
exchange. 1 

A public exhibition of the working of this system was given 
on May 17, 1877, and on the following day the Boston Evening Tran- 
script reported that ' conversation was carried on between the several 
points connected with perfect ease.' On May 15 Mr. Holmes 
ordered from Mr. Williams a new six-plug switch, that the manipula- 
tion of the lines might be more easily accomplished than his then 
existing burglar alarm apparatus permitted. Additional subscribers 
were connected at this time, and Mr. Holmes's appreciation of the 
utility of the system is evidenced by his desire to obtain the central 
office rights, as indicated in the following correspondence : 

Boston, July 19, 1877. 

DEAR SIR, I understand from my conversations etc. with you 
that I am to have the exclusive right of the use of your telephones 
for the city of Boston for all Central Office purposes. 

The plans most definitely mentioned so far, being a system of 
running a single and exclusive wire from a subscriber's business 
place or residence to our central office for the purpose of putting 
said subscriber in a direct speaking communication with any other 
subscriber to our central office. 

Another being a plan now being done by the District Telegraph 
Company, that is : our Subscribers, a number being in the same 
circuit or on the same wire, can telephone the central office for a 
messenger, or call to any express office in the city also Railroad, 
Telegraph and Newspaper offices, Mercantile Agencies etc., providing 
a demand is made for the connections, and also any other public 
points that our subscribers may demand as being necessary. 

The above plans, yourself and I have thought of, perhaps the 

most, but as I understand it I am to have the use of the telephone 

for all ceniral office purposes which we may now have in mind, 

or which miv in future show themselves to us through the experi- 

1 Boston Electrical Handbook (1904), p. 126. 


ence and acquaintance which we may gather from the constant 
use of the telephone in the business which we now propose to 
work up. . . . 

Yours very respectfully, 

Mr. Hubbard's reply was as follows : 

Cambridge, August 10, 1877. 

E. T. HOLMES, Esq., 

DEAR SIR, It is understood that you are to have the exclusive 
right for the use of the telephone in the city of Boston and within 
a circuit of ten miles for central office purposes viz. for the use of 
Telephone in circuits connecting a central office with houses, offices, 
stores and other buildings. 

The Bell Telephone Company will furnish the telephones for 
$10 a year payable quarterly in advance. 

The several cities within your territory are to be provided with 
similar circuits to central offices communicating with the central 
office in Boston within two years. If not the Bell Telephone 
Company shall have the right to authorise other parties to establish 
such circuits, these several central offices connecting with the Boston 
central office in terms to be agreed upon or fixed by a referee. 

The right hereby granted to you to be exclusive provided you 
serve the public promptly and faithfully, constructing new or 
extending the old circuits as the demand increases, performing the 
entire business to the satisfaction of the Bell Telephone Company. 

The Bell Telephone Company reserves the right to purchase 
these lines with the goodwill of the business at an appraised 
valuation, but not exceeding the actual cost of the lines, or for the 
goodwill not exceeding $5000 if taken within three years from the 
date hereby given, 

I am, Yours Truly, 


The exchange system so clearly in the minds of the writers 
was not at once developed by Holmes. The business requirement 
of the public which seemed to offer the more immediate prospect 
of success was that of connecting subscribers to a central office to 
which they might transmit orders for a General Express Agency, 
these orders being re-transmitted from the central office. The 
pecuniary value of such a service was more readily demonstrable, 
and it was along such lines that the telephone was first developed in 
Boston. ' Experience and acquaintance,' which might be gathered 
' from the constant use of the telephone,' was not needed by those 
who proposed to introduce the central office system, but rather by 
the public whose patronage was essential for its remunerative 


This experience was being gathered in other directions. Isaac 
D. Smith was proprietor of the Capitol Avenue Drug Store at 
Hartford, Connecticut, and in addition to being a chemist took great 
interest in philosophical and electrical discoveries. For the con- 
venience of his business he had constructed in May 1877 a line 
between his drug store and the office or surgery of Dr. James 
Campbell, 28 Buckingham Street. At each end of this line was 
a bell and a push-button. Signals were exchanged by means of a 
prearranged code. In July 1877 a second line was constructed to 
Boardman's Livery Stable, 104 Main Street, where Dr. Campbell's 
horse was kept. When the doctor required his horse he would, for 
example, strike two on the bell at the Drug Store and the attendant 
would repeat the signal to the stable. Mr. Smith's scientific 
interests led to his being familiar with the fact of the introduction 
of the telephone, and in July 1877 he obtained the agency of the 
New England Telephone Company, when telephones were added 
to the lines, first to the doctor's house and later to the stable also. 

The method of repeating the calls was at first adopted with the 
telephones in the same way as with the bells, but after a few days, 
and at the instance of Dr. Campbell, Mr. Smith devised an arrange- 
ment by means of which the lines might be connected together. 
The successful accomplishment of this arrangement led to other 
doctors being added to the system, and on August 17, 1877, the 
Hartford Courant contained the following announcement : 

At the regular meeting of the allopathic physicians on Monday 
evening experiments were successsfully tried with telephones, and 
it was proposed to have a system of inter-communication between 
the doctors established by means of the new invention so that by 
reporting to a central office at the Capitol Avenue Drug Store they 
can readily exchange views between office and office. 

The results were so far encouraging as to induce the chemist to 
enlarge the field of his electrical operations, and on October 8, 1877, 
he advertised in the Hartford Courant, under the heading ' Profe'ssor 
Bell's Telephone,' that he was prepared to build and equip telephone 
lines at moderate rates. By the month of November 1877 there 
were seventeen subscribers with telephonic connection, though not 
all connected with the central office. The line of one was extended 
to another, and various branches thus existed with several stations 
on them, but all able to inter-communicate on request, though some 
required to do so by various stages. 

The practical advantages of Mr. Smith's central office system 
and the utility of the telephone as a means of prompt communica- 
tion were shown on the occasion of a railway accident in the 
neighbourhood during the month of January 1878. This circum- 


stance is referred to in the following extract from a Boston newspaper 
of the period : 

A striking illustration of the use of Professor Bell's Telephone 
in a case of emergency was furnished by the recent terrible disaster 
to the Moody and Sankey excursion train on the Shore Line [? Con- 
necticut Western] Road. 

The first news came in the shape of a message to the Western 
Union Office in Hartford from Tariffville, stating that a fearful 
accident had happened at that place, and asking that as many 
surgeons as could be obtained be sent at once. 

The operator at Hartford immediately telegraphed to Isaac 
D. Smith's Capitol Avenue drug store, and the night clerk was 
awakened and told what was wanted. 

The drug store is connected by telephone, so that twenty-one 
different physicians could be summoned, which was done, and at 
the same time the clerk telephoned to a livery stable for an express 
wagon, into which on its arrival was packed bandages, morphine, 
chloroform, and whatever was thought would be needed. 

Meanwhile the doctors, as instructed, went down to the Hartford 

Supt. J. T. McManus of the Hartford Providence and Fishkille 
Road had also been aroused, none of the C. W. people being 
handy, and he had at once gone to work and gotten out an engine 
and some wrecking cars, which he had all ready on the arrival of 
the physicians. Extra passenger cars were also added, the bandages, 
drugs, etc. were taken aboard, and about a dozen doctors started 
for Tariff ville.i 

This was the earliest instance of the benefits that the telephone 
and the exchange system were capable of rendering in a great 
catastrophe, and its publication in the press served to arouse public 
interest in the invention and the application. To the press indeed 
the telephone has been much indebted from its earliest days for 
recording its capabilities. And to a member of the press credit 
must also be given for a recognition of the value of the exchange 
service in advance of its accomplishment. 

Mr. Ponton, a reporter for the Titusville Herald, found in the 
telephone a suitable subject for what is known in press circles as 
' copy.' He was present at many of the earlier exhibitions and 
lectures given by Bell. The Titusville Herald of October 22, 
1876, had a paragraph on the telephone. The Salem experiments 
of February 12 and March 5, 1877, were also described in that paper. 
Mr. Ponton was so much impressed with the utility of the telephone 
that he became one of the earliest agents of the Bell Telephone 
Company. On August 7, 1877, he wrote to Mr. Hubbard : 

1 Semi-Weekly Advertiser, Boston, January 25, 1878. 


There is no difficulty whatever in starting the central system here 
and spreading it all over the oil region, and I will commence active 
operations at once so soon as our agreement is definitely concluded. 

On August 19, 1877, he wrote at length on the subject. He 
referred to the question of interesting capitalists, made certain 
estimates of costs showing 

that the outlay of capital, even for towns where the wire has not 
to be sunk underground, is four times greater than that of the 
telephones themselves, and this is a very low estimate it may 
in some instances be five times. 

The amount estimated for the telephones was $10 each. He also 
prepared a scheme of rentals, which should be a source of income 
to the Bell Telephone Company as well as the capitalists concerned, 
and concluded 

All I want is to have the exclusive agency or right to use in places 
only where a certain amount of capital is expended in the formation 
of the Central System as above described, and leave it to you to 
designate the counties in which it shall be done. 

On October 27, 1877, he forwarded a copy of a circular headed 
' Ponton's Telephone Central Service,' and stating 

This service is original with the subscriber (Ponton) and was 
proposed by him to the members of the Bell Telephone Company 
before any private lines were built, and it met their hearty approval. 
It was too early at that time, however, to start the system, as the 
public required further evidence that the Telephone would do its 

The system is extremely simple. All parties who wish to adopt 
it must have a separate wire from their house, office, factory, hotel, 
store, bank or restaurant to a central switch room, where any one 
wire can instantaneously be connected with any other wire. 
Supposing that one hundred persons adopt this system, and that 
the average length of each wire is half a mile, it would give each 
person the privilege of using fifty miles of wire at a less cost than 
it could be done with only one mile in the private line. 

All the advantages of the telephone exchange system are clearly 
set forth, and the circular contains a list of occupations which might 
have been left to the imagination of the reader. ' In domestic 
life,' it states, ' the telephone can put the user in instant communica- 
tion with the grocer, butcher, baker,' and one hundred and seventy- 
six separately stated other occupations, followed by the general 
phrase, 'and other places and persons too numerous to mention.' 


Unhappily the work of this enterprising journalist was 
unproductive, for he wrote under date of January 3, 1878, that 
he had failed to arouse sufficient interest to establish his 

The scepticism was not confined to the public. On August 17, 
1877, the Glasgow Herald stated that 

In America, at Boston and Cleveland, Ohio, dwelling houses 
and places of business are connected with the telephone with a 
central office. Merchants talk with each other without the inter- 
vention of a third party, and therefore in perfect secrecy. Brown 
telephones his wife that he is bringing Jones and Murphy to dinner 
at five o'clock, and straightway Mrs. Brown directs the central 
office to give her the butcher, and a joint is immediately, ordered 
by means of the telephone. It is further intended that the cities 
of the New World be placed in telephonic communication with a 
Central Office, where a merchant can buy the use of the line for 
ten or twenty minutes, as the case might be. For example, a 
merchant may write to another friend miles away ' to meet him at 
the telephone exchange at 12 o'clock sharp,' and next day they 
converse as freely as if they were in private one with the other. 

There is good reason to believe that this statement was the 
echo of Mr. Ponton's or some other press representative's ' intelligent 
anticipation of events ' rather than a record of facts, so that there 
was some justification for the London Telegraphic Journal of 
October 15, 1877, preceding its quotation from the Herald with the 
observation : 

Like other marvellous things, the telephone seems to have 
established quite a literature of its own. The comic papers have 
employed it as a vehicle for their wit, and especially for that of 
a rather far-fetched description. Poets have eagerly welcomed it 
as a new image, and there have not been wanting preachers who 
have hailed it as a new symbol. It has been the theme of a great 
deal of amusing speculation, of which it is difficult to distinguish 
jest from earnest. 

The idea of the exchange seems to have occurred to many men 
in widely separated places during the latter half of 1877. Some 
of them were familiar with telegraphic systems which conveyed 
information to the public, such as that which is known in England 
as the Exchange Telegraph Service and in the United States as the 
Gold and Stock Telegraph, or of that known in both as the District 
Messenger Service. The Gold and Stock telegraphed to its sub- 
scribers the premium rate for Gold and the prices for Stocks and 
Shares. The District Messengers were summoned by a subscriber 


pulling a lever which sent over a common line a series of electric 
impulses recorded at the Central Office and identifying that sub- 

Such services as these had been used by a limited number of 
business men, and had moreover educated some other men in tele- 
graphic work, so that they were more ready than the totally inex- 
perienced to see advantages in any new development of telegraphic 
communication. The telephone was such a new development. 
No one name is on record as the originator of the exchange system, 
probably for the reason that no one is entitled to such credit. There 
were many to whom such an application occurred simultaneously. 
But Bell and his associates were fully alive to its value, offered it 
every encouragement, and proceeded with its organisation in a 
masterly way. 

The telephone was already an aid to commerce, it enabled distant 
converse between fixed points over distances ' not exceeding twenty 
miles,' with intermediate listeners if desired. Easy conversation 
was replacing cumbrous messages, and confidence that plant of 
proverbially slow growth was replacing the attitude of incredulity 
with which the commercial mind regarded the telephone on its 



To the present generation familiarity in the use of the telephone 
has dulled the sentiment of wonder once existing that speech 
should be transmitted or that means should be found for diverting 
that speech from line to line at the speaker's desire. So promptly 
was science applied to commercial and social uses ; so generally, 
after a time, was the application taken advantage of, that the wonder 
has evaporated and the user only become the more exacting in his 

Electrical communication was the function of the telegraph. 
Electrical intercommunication of anything like a general character 
was reserved for the telephone, although efforts were made to 
utilise the telegraph for what are now known as exchange purposes. 

The earliest suggestion so far discovered for the interconnection 
of telegraphic lines for public use through a central office or central 
offices is that of Frangois Marcelin Aristide Dumont of Paris in the 
Republic of France, engineer. He took out a British patent 
in 1851, the year in which the then recent advances in science and 
manufactures were signalised by the Great Exhibition in London. 
Dumont's patent is numbered 13,497. He states that his invention 
consists in : - 

First. The employment of the electric telegraph for the con- 
veyance of intelligence in the interior of large towns, as London or 
Paris, by a particular combination of electric wires, and by a new 
system of placing and fixing such wires. . . . The means I employ 
for effecting a communication in towns consists in establishing, 
in any town to which my system may be applied, a number of 
stations, and connecting these stations to one another, or to the 
house of each subscriber, by the particular arrangement of electric 
wires hereinafter described. 

Fig. i, sheet i, of my drawings represents a central station, 
to which each of the small stations is attached by one or several 
wires which only communicate with it. ' O ' is the central station. 



Nos. I, 2, 3, 4, 5, 6, 7, 8 and the following are corresponding small 
stations. Each corresponding station is for a certain number of 
houses or subscribers. Each house has an electric wire connecting 
it to the station, and a double electric apparatus, one at the sub- 
scriber's house and the other at the neighbouring station to which 
it is connected. The houses are telegraphically numbered from 
one to any number. Thus, suppose the house number 3, con- 
nected to station No. I, and the occupant of house No. 3 is desirous 
to communicate with the house 287, connected with the station 12, 
the subscriber of the house No. 3 signalises to station i, the number 
287. The clerk of the station No. I then directs the central station 
O to place the wire i.O in communication with the wire 0.12. When 
this is done the clerk signalises to station 12 the No. 287. The 
clerk at station 12 then connects wires O.I2 with the wire 12. 287. 

This accomplished, the clerk of station No. i finally connects 
the wires 3 and i.O and a direct and intermediate communication 
is then established between the houses 3 and 287, and both sub- 
scribers may correspond privately. 1 

Figure 25 is a reproduction of fig. i of Dumont's specification, 
and shows quite clearly that he intended to protect the exchange 
system. Though he illustrated telegraph instruments in his patent, 
the wording of the first claim would have entitled him to regard the 
use of telephones as coming within it. The claim is as follows : 

First, the system or mode of applying electricity to the intercom- 
munication of large towns, and the interchange of communication 
between the inhabitants of such towns as hereinbefore described. 1 

Dumont was born at Crest (Department of the Drome) in 1819. 
Educated at the Ecole Polytechnique he entered, in 1838, the 
Ecole des Ponts et Chaussees and became engineer in chief of the 
second class. According to the Grand Dictionnaire Universel du 
igeme siecle, vol. 6, his activities covered a wide field in engineering, 
for he published a number of works, including an essay on the 
embankment and canalisation of the Rhone (1840), ' Paris port 
de mer ' (1863), and several works on kindred subjects; but ' La 
reforme Administrative et les Telegraphes electriques ' (1849) 
is the only electrical work which is recorded. A French patent, 
No. 10.439, dated September 1850, intituled ' Perfectionnements et 
Applications de la Telegraphic electrique ' has an addition of 
November 30, 1850, which is apparently the equivalent of the English 
patent of 1851. In view of this patent cordial agreement may be 
expressed with the remark of the editor of the Grand Dictionnaire 
Universel, who says : ' The writings of this engineer scientist are, 

1 British specification, No. 13,497 of 1851. 


remarkable for a great breadth of view and a vivid comprehension 
of the wants of modern civilisation.' 

Dumont's patent was presumably known at the time of its 

FIG. 25. Dumont's Patent of 1851 for Telegraph Exchange (first figure). 

publication to those interested in telegraphy, but I have found no 
reference to it in telegraphic literature, and any influence that it 
may have had on subsequent work in the same direction is, 
so far as I have been able to learn, unrecorded. 


George B. Prescott in his ' History, Theory, and Practice of 
the Electric Telegraph,' published in 1866, quotes an article from 
All the Year Round, 1859. In the course of this article it is said : 

The industrial spiders have long since formed themselves into a 
commercial company called the London District Telegraph Com- 
pany (limited), and they have silently but effectively spun their 
trading web. 

One hundred and sixty miles of wire are now fixed along parapets, 
through trees, over garrets, round chimney pots, and across the 
roads, on the southern side of the river, and the other one hundred 
and twenty required miles will soon be fixed in the same manner on 
the northern side. . . . Other labour will be required to bring down 
the mysterious strings so that every one may be able to move 
the living puppets, from station to station, from Highgate to 
Peckham, from Hammersmith to Bow. Some of these strings, 
perhaps to the number of ten, will drop into district stations, offices 
that will act as centres of particular divisions ; others (perhaps to 
the number of a hundred) will drop into familiar shops and trading 
places, amongst pickle jars of the oilman, the tarts of the pastry 
cook, the sugar casks of the grocer, the beer barrels of the publican, 
and the physic bottles of the dispensing chemist. . . . 

The great centre of all this system is in Lothbury, London, 
where a graceful school of about sixty young ladies are even now 
learning the mysteries of the old railway telegraph signals. 

Whatever machines may be used at the central and district 
stations, it is certain that the sub-district or shop stations will 
require something exceedingly simple and convenient. 1 

The article goes on to describe Wheatstone's ABC telegraph 
instrument, and concludes : ' Upon the adoption of some such 
apparatus as this most probably upon this particular machine 
will depend the success of the London District Telegraph Company.' 

Our present familiarity with the terms ' Central ' and ' District ' 
would make it easy to read into such a description as the foregoing 
a reference to a central office or exchange system, but the prospec- 
tus of the Company amongst the papers of Sir Charles Bright in the 
library of the Institution of Electrical Engineers shows that at the 
' Central ' in Lothbury or at the ' District ' offices there was no 
switching apparatus. The description relates simply to a message 
sending arrangement, and it was apparently contemplated to use 
' familiar shops ' for the purpose of transmitting messages, so that 
the expense of a special staff might be saved just as post offices are 

1 All the Year Round (conducted by Charles Dickens), November 26, 
1859. No. 31, p. 106. The article is entitled ' House-Top Telegraphs/ and 
appeared in the issue of the magazine which contained the final chapter of 
' A Tale of Two Cities.' 


combined with trading establishments at the present day. The 
use of overhouse wires was put forward as an important feature in 
the prospectus. 

The Telegraphic Journal of May 7, 1864 (p. 218), quotes from 
Chambers's Journal an article on the District Telegraph system 
under the title of ' The Domestic Telegraph.' The principal office 
at this time appears to have been in Cannon Street. That no 
intercommunicating arrangement was provided for in the system 
is clear from the following paragraph : 

Should the message be from one out-station to another the clerk 
at Cannon Street copies the message, and delivers it into the hands 
of the person who works the particular instrument to which the 
office for which the message is intended is connected. The clerk 
there again copies the message, and delivers it to a boy, who carries 
it to its final destination. 

Yet Dumont's patent had been published thirteen years before, 
and the utility of intercommunicating did not escape the writer 
of the article, as a later paragraph shows : 

In an upstairs room of the office in Cannon Street there is a kind 
of cupboard, upon opening the doors of which a mass of wires and 
screws, bolts and fastenings, are exposed to view. These are all 
lettered or numbered, and a strange mystery exists with regard to 
them ; they are really the junctions between various station-wires, 
and thus two extreme stations might at once be placed in connection 
with each other. We will suppose that B represents the termination 
of the wires that lead from Blackheath to Cannon Street, by which 
wires the messages are sent, and C the termination of the Camden 
Town wires. Now B and C might be a yard apart in the cupboard, 
and it would merely be necessary to join these two by means of 
a wire, in order to allow Camden Town to talk to Blackheath, or 
vice versa. [To ' talk ' or to ' speak ' was telegraphese for sending 

Incidentally it may be remarked that the district system would 
appear to have been the pioneer of cheap telegraphy. Reference 
is made to the fact that the public could now procure one hundred 
of the company's stamps for one pound. A stamp placed on a 
written message, which might be enclosed in an envelope and sent 
to the nearest station, would ensure the transmission of the message ; 
thus ' the actual price of a message of fifteen words is rather under 
twopence-halfpenny . ' 

According to the report of the company for the half year ending 
June 1864, 152,795 messages were sent and 4,802 los. received 
for them, so that the average price per message was approximately 


j\d. The receipts fell short of the expenditure by 60 izs. for the 
six months, an improvement upon previous years which gave the 
shareholders some hopes ; but there was considerable disappointment 
that the public did not offer greater support to the undertaking. 
This was attributed by one of the shareholders to the fact that the 
company had but fulfilled one half of what they undertook to do, 
and what they hoped still to do. He found that in many cases 
there was such a delay in the delivery of messages that they could 
be as speedily delivered by messengers. Many complaints had also 
been made of the inaccuracy of the messages, and some respecting 
the charges. 

The experiences of the London District Telegraph Co. were not 
encouraging. In spite of ' publicity methods ' which a later age 
might envy, the directors were, after four years' efforts, unable 
to develop a paying traffic in local telegrams over the London 
area, though they apparently served a useful purpose as collectors 
for the long distance companies, since in addition to the amounts 
already mentioned they received 3,202, which was paid away to 
other companies. Whilst the local systems were much developed in 
later years, especially after their acquisition by the Post Office, there 
was never a popular local use of the telegraph. The introduction 
of pneumatic tubes facilitated the transmission of messages to some 
extent, but the message was still only a message. Local inter- 
communication on a large scale had to wait until the trans- 
mission of speech became practical, and the exchange system 
developed, for the exchange system did not originate with the 

Telegraphic exchange systems were established both in Great 
Britain and the United States many years before the telephone was 
invented. The information available regarding these early tele- 
graphic systems is very meagre, and with a view to ascertain the facts 
so far as possible regarding the British systems, I inquired of Mr 
A. M. J. Ogilvie, C.B., one of the Secretaries of the Post Office, 
who has kindly furnished me with information obtained from 
officials of the department who were personally concerned with 
their working. That relating to Newcastle is furnished by Mr. 
Colin Brodie and Mr. Shadforth, from whose recollections the 
following is compiled : 

The origin of the exchange system was due to the lack of postal 
and telegraphic facilities in the district, so that it was necessary to 
have a messenger attached to each colliery or works for the purpose 
of conveying correspondence to and from the different agents and 
business firms. This arrangement was naturally slow, as fairly long 
distances had sometimes to be travelled. It was therefore a relief 
to some of the larger firms when the Universal Private Telegraph 


Company 1 offered to erect lines at a certain rental and agreed to 
dispose of the telegrams received from the subscribers for delivery 
or transmission. This was in 1865, and the service began with 
three or four renters. By 1870, when the telegraphs were taken 
over by the Post Office, the number had slightly increased. In 1872 
Mr. Colin Brodie sought to develop the system, and succeeded so well 
that a switch for sixty lines was fixed and brought into operation 
in 1873. He invited firms who rented private lines to extend to the 
Post Office and thereby secure the means of speaking to each other. 
There were about forty renters and twenty post offices connected 
to the switch when, in 1881, the subscribers were invited to adopt 
telephones instead. Most of them agreed, but some retained the 
ABC sets in case of failure. These were, however, soon discarded. 
The substituted telephone service was started in 1882. 

The rentals for the Newcastle Telegraph Exchange were 
calculated as follows : 

Wires. j per mile, with a minimum of \ mile for those 
who were renters of other lines, and I mile for independent con- 
nections with the exchange. 

Instruments. A B C sets, 6 each ; extra bells, T each ; 
switches and reversions, IDS. ; clerk's services, 5 55 ; apparatus 
at Post Office was 3 or 3 los. per annum. 

In Glasgow also the Universal Private Telegraph Company 
carried on business, but apparently had not installed any exchange 
system prior to their lines being taken over by the Post Office in 
1870. Mr. D. Campbell, Assistant Postmaster and late Super- 
intendent of Telegraphs at Glasgow, relates that intercommunication 
was set up by the Post Office on an ' Umschalter ' switch. There 
were at one time as many as eighty renters. The work done 
included not only switching for intercommunication purposes but 
also the reception of telegrams for local delivery and outward 
transmission, and the delivery by private wire of inward messages. 
It will certainly be a matter for general surprise that this telegraphic 
' exchange ' still exists (January 1914), though in a very attenuated 
shape. It is now part of an A B C Concentrator, and of the four lines 
connected to it three are those of orginal renters. 

In addition to those at Newcastle and Glasgow a telegraphic 
exchange system also existed in Bradford in the pre-telephone era. 
The formation of similar systems at a later date is indicative of a 
want of confidence in telephonic communication, but such telegraphic 

1 The Universal Private Telegraph Co. was incorporated by Act of Parlia- 
ment, 24 & 25 Viet. (Royal Assent June 7, 1861). It was intended to establish 
private telegraphic lines using \\heatstone ABC instruments. There was 
no mention in its prospectus of an exchange or switching system which 
apparently developed from local circumstances after a number of private 
lines had been put into operation. 

G 2 


exchanges were introduced between 1878 and 1881 in Swansea, West 
Hartlepool, Middlesbrough, Paisley, Leeds, Darlington, Stockton-on- 
Tees, Hull, Sunderland, Bristol, Barnsley, and Workington. 1 

Sir John Gavey, C.B., who was in charge of the district at the 
time, has kindly furnished me with the following particulars of the 
Swansea system. 

P~rom time to time colliery proprietors and others entered into 
contracts with the Post Office for the provision of private wires 
worked with Wheatstone ABC instruments and designed for the 
dispatch and receipt of telegrams, &c. In 1878 arrangements 
were made to provide intercommunication between these lines, and 
this was done in the following manner : 

Each line terminated at the Post Office on a Wheatstone ABC 
receiver and through an Umschalter switch to earth. A pole changer 
or reversing switch was in circuit with each line. A call for the Post 
Office, which was indicated by a repetition of the code, was answered 
by the operator, who switched into circuit a complete ABC set 
and dealt with the call. A call for another subscriber was switched 
through to the latter on the Umschalter, the reversing switch being 
turned so that the two instruments should be in unison. The 
ABC receivers on the intercommunication switch were always 
in circuit, so the operator was cognisant of what passed and knew 
when to sever the connection. There were ten subscribers, and the 
service was opened on October 7, 1878. 

On March 23, 1881, the ABC instruments were replaced by tele- 
phones, and the Post Office telephone exchange at Swansea started. 2 

Similar telegraphic exchange systems were in existence at New 
York and Philadelphia. For the following particulars regarding 
these systems, also compiled from evidence furnished by partici- 
pators in the service, I am indebted to Mr. Lockwood. 

The Gold and Stock Telegraph Company established in New 
York in the year 1869 a telegraphic exchange system in connection 
with the Clearing House, and having direct lines to various banks. 
The Central Office was situated at the Clearing House at the corner 
of Wall Street and William Street. The business of that central 
office was chiefly to report to each bank in the system by means of 
printing instruments its daily debit or credit balance, and to repeat 
to any bank any message received for it by wire from any other 
bank. But the wires of any two banks desiring direct communica- 

1 General Webber relates that the Post Office used a switch at Colchester 
for the ordinary business between several small towns in that district (Journal 
of the Society of Arts, April 28, 1882, xxx. 609.) 

* Technical descriptions and illustrations of these telegraphic systems 
are given by Mr. Purves, M.I.E.E. in Telegraph Switching Systems (Alabaster, 
Gatehouse & Co. 1902). 


tion were also connected together at a switch. This direct con- 
necting system was not much used at first. In 1871 the Central 
Office was removed to the general office of the Gold and Stock 
Company at 61 (and later to 197) Broadway, an instrument being 
left at the Clearing House, which then became equivalent to a 
subscriber instead of being the exchange. An operator was sent 
daily to the Clearing House, who reported to the Central Office at 
61 Broadway each bank's balance, which was then repeated to the 
bank by an operator and not by connecting the various bank lines 
with the Clearing House lines ; but from 1871, by means of a suitable 
battery and the regular switching devices, any bank was on demand 
connected with any other bank or with the Clearing House. All 
the wires came to the switch (a Jones Lock switch) through a relay, 
and these relays, one or both, were left in circuit, and of course 
rattled and clicked all the time the two banks were working together 
by their printing instruments. When the subscribers required to 
be disconnected they made three long pulsations as a signal, but 
this signal was frequently forgotten to be given, so that when the 
relays were silent for a considerable time they were disconnected 
without signal. About twenty-five banks and two private lines 
were connected to this system, which continued in existence until 
i88o. 1 

A telegraphic exchange of a similar character was established 
at Philadelphia, the following particulars of which are taken from 
a letter written by Mr. Bentley, the proprietor, to Mr. Theo. N. 
Vail on December 12, 1881 : 

I first established a system of connecting private telegraph 
wires through a central office in this city in the year 1867 (in August), 
as I find by my records. These lines were Morse telegraph wires, 
and the subscribers to my plan of intercommunication included 
the largest houses in the city. 

In September 1871 I arranged a central office for intercom- 
munication between some twenty odd of the banks and bankers 
of this city, having said central office in a room adjoining the clearing 
house. . . . Within the five years following my establishment of 
a central office for private wires my subscribers increased to some- 
where about fifty. 

The first telephones We ever worked, so far as I find any record, 
were four, which were sent to myself personally by Mr. G. Hubbard, 
or by his direction. My books show that I received them by express 
August 31, 1877, worked them around on various short wires to test 
them, and on September ir following we connected the office 
of H. D. Davis, 3rd and Walnut Streets, also in the mean time the 

1 From statement of Mr. G. F. Wiley, an officer of the Gold and Stock 
Telegraph Company, in the possession of Mr. T. D. Lockwood. 


Stock Exchange and my office. These were of course experi- 
mental tests. I give this to you for what it is worth. Finally, Mr. 
Cornish was directed to get the Phones from me and that ended that 

The first telephones we rented for pay from this office was on 
February 14, 1878. 

Another system of the same character was that established 
in New York by the Law Telegraph Company. 

The origin of this system was related to me some years ago 
by its originator, Mr. William A. Childs. 

In the spring of 1874 [he said] I proposed to establish for lawyers 
in New York City a circuit wire system with tickers or printing 
telegraph instruments a service similar to that given brokers and 
bankers by the Gold and Stock Telegraph. My idea was to connect 
the lawyers with the various courts so that they could get each day 
the calendars for the next day, the decisions of the Judges, and 
Court news generally. I went among the lawyers and solicited 
them with a subscription paper, but they said, ' That will be no use 
to us.' After canvassing for a week or more with no encouragement 
a lawyer one day said to me, ' There is a great deal of intercom- 
munication amongst lawyers. What you propose is no use; but 
if you had a system of telegraphs by which I could telegraph to 
any other lawyer who had one, that is something I would pay for.' 
Another lawyer said my plan was of no use. He could get all 
that I proposed in the morning or evening papers, but a 
means of direct telegraphic communication would be of value. 
I then conceived the idea of the establishment of a central office 
with a system of signals whereby if Smith wanted to talk with 
Jones he could signal the Central Office and have his wire switched 
on to Jones's wire, and so on. Then I considered the possibilities 
of getting the wires up and what instruments to use. We adopted 
the dial instrument similar to Wheatstone's ABC made by 
Chester, of New York. Then I printed a new prospectus setting 
forth the new plan and commenced to work a few weeks after. 
The plan went ' like hot cakes,' subscribers putting their names 
down readily. There was one central office with a single wire to 
each subscriber and an individual annunciator on each line. The 
annunciator was a bell. The number required was indicated by the 

strokes on the bell of the subscriber giving the call as for 31. 

Sixty bells were placed in a space of, say, four feet long by three feet 
high. In front of these bells stood about three operators. On the 
hammer of the bell was a tag containing the subscriber's number. 
When a subscriber called the Central Office the key which he used 
for the purpose automatically shunted a large amount of resistance 
normally in the circuit and thereby gave the battery located at 
the Central Office sufficient power to move the Central Office bell 
which otherwise it did not have. A signal to disconnect was made 


with the same key as the call, hence resistance was taken out and 
the Central Office bell rang. In each line there was normally 
connected twelve cells of gravity battery at the Central Office. 
When we coupled two subscribers together there were thus twenty- 
four cells in the combined circuit, and in coupling them we neces- 
sarily had to reverse the direction of twelve of those cells which 
required a switch properly constructed for the purpose. 

The bell described by Mr. Childs as the first annunciator was 
later superseded by a Morse Sounder upon a separate line a call 

References are made to the Law Telegraph system in ' Tele- 
grapher,' April 10, 1875, and May I, 1875. The following is taken 
from a contemporary notice in a New York paper : 

Electricity has now become identified with the interests of the 
whole world, from the mightiest ramifications to the demand of 
local business necessities, and supplies a means of direct and private 
communication between lawyer and client, merchants, manufac- 
turers, &c., and their connections, indeed, throughout the city, 
country, and world at large. Most especially is this effectually 
secured by the Law Telegraph Company of 145 Fulton Street and 
261 Broadway, New York, which was organised in the year 1874, 
and was established for the special purpose of members of the legal 
profession having direct and immediate communication with their 
associates, clients, and personal offices, thus entirely dispensing 
with the uncertain and tardy system of conveying information by 

After a general description of the ' instruments ' and ' central 
office/ it is stated that 'Arrangements are now being made for 
direct connection with the Western Union Telegraph Company 
whereby messages can be sent to any part of the world by sub- 
scribers' own instruments,' a statement which clearly must not 
be taken too literally. We can form an idea of the subscription 
rates as well as learn that the Company did not stand alone by 
what follows : 

The subscription price of the instrument, compared with that 
of other companies, is in itself an important recommendation ; for 
instance, brokers pay such companies at the rate of $25 per month, 
cotton merchants $30 per month, and parties having private lines 
are charged from $300 to $1,200 per year, while the charges of the 
Law Telegraph Company are from 20 per cent, to 30 per cent 
cheaper. . . . Since the establishment of the Company the number 
of subscribers has been doubled, and some one hundred and 
twenty instruments are now in use. 


The names are given of ' a few of the prominent houses and firms 
who have ' long used them/ and the article concludes with the 
following : 

Setting the average cost at $5 per week for an ordinary sub- 
scription, all parties must perceive how entirely inadequate it is to 
the general value derived from the use of the machine, and the 
attention of the business and professional community should be 
closely interested in the system, indeed no extensive establishments 
should fail to avail themselves of it. 

As we have seen, telegraphic intercommunication was suggested 
in 1851 by Dumont, whose patent is quoted above. It existed in 
Newcastle-on-Tyne in 1865 and in Philadelphia in 1867 ; amongst 
the bankers of New York in 1869, and the lawyers of the same 
city in 1874. 

But so long as words had to be transmitted in dots and dashes 
or spelt out by letters upon a dial, there was little likelihood of the 
extensive use of an exchange system. With the introduction of 
an instrument that talked, the situation was completely changed. 
There was no need then for dependence on skilled workers. Com- 
munications were no longer restricted to mere messages, but were 
expanded to the very different condition of conversations. The 
enormous gap that was thus bridged is not recognised so generally 
as it should be. Question and answer may be sent by telegraph, 
but they must always lack the spontaneity of a conversation such 
as the telephone affords. This fact alone suffices to explain the 
little use made of the exchange idea in conjunction with the tele- 
graph, and to indicate how the invention of a talking instrument 
changed the situation. The inventor of that instrument was 
amongst the first to recognise the advantages which the exchange 
system would confer. In the ' Boston Electrical Handbook ' it is 
related that at the New York lectures on the evenings of May 
17, 18, and 19, 1877, Bell outlined and eloquently advocated the 
proposed use of the telephone exchange yet to be developed. 1 
Mr. T. D. Lockwood was the writer of this portion of the Hand- 
book. He was present at the lectures and records his own 

On August 4, 1877, Bell sailed for Europe and did not return 
until November 10, 1878.2 The development of the business in 
the United States consequently devolved upon his associates, but 
he delivered lectures and gave demonstrations of the use of the 
telephone at the Society of Arts and the Society of Telegraph 
Engineers in London, and before the British Association at Plymouth. 

1 Boston Electrical Handbook, p. 126. * Deposition, p. 178. 


Negotiations were in progress for the acquisition of the English 
patents, and in connection with these negotiations Bell prepared 
a statement, printed below. In a letter to E. J. Hall dated 
June 8, 1903, enclosing copy of the statement, Bell says : 

I may briefly recite the circumstances under which this paper 
was written : 

I was married on July n, 1877, and went abroad very shortly 
afterwards and remained continuously abroad for about one year. 
Colonel Reynolds, of Providence, R.I., undertook to organise in 
England a telephone company to work my invention there. He 
was at work upon this in the winter of 1877 and spring of 1878, and 
early in 1878 Colonel Reynolds requested me to write an address 
to the capitalists he had succeeded in interesting, setting forth the 
advantages of the telephone. This I did in a paper dated from 
Kensington, March 25, 1878, addressed to ' The Capitalists of the 
Electric Telephone Company.' I remember that this paper was 
printed as a kind of prospectus, and a copy given to the gentlemen 
concerned and others interested. I beg to enclose a copy of this 
document. It expresses, of course, the ideas of the development 
of the telephone I had in mind in the autumn of 1877 when I went 
to England. 

Kensington, March 25, 1878. 

To the Capitalists of the Electric Telephone Company. 

GENTLEMEN, It has been suggested that at this, our first 
meeting, I should lay before you a few ideas concerning the future 
of the Electric Telephone, together with any suggestions that occur 
to me in regard to the best mode of introducing the instrument to 
the public. 

The telephone may be briefly described as an electrical con- 
trivance for reproducing in distant places the tones and articulation 
of a speaker's voice, so that conversation can be carried on by word 
of mouth, between persons in different rooms, in different streets, 
or in different towns. 

The great advantage it possesses over every other form of 
electrical apparatus consists in the fact that it requires no skill 
to operate the instrument. All other telegraph machines produce 
signals which require to be translated by experts, and such instru- 
ments are therefore extremely limited in their application, but the 
telephone actually speaks, and for this reason it can be utilised for 
nearly every purpose for which speech is employed. 

The chief obstacle to the universal use of electricity as a means 
of communication between distant points has been the skill required 
to operate telegraphic instruments. The invention of automatic 
printing, telegraphic dial instruments, &c., has materially reduced 
the amount of skill required, but has introduced a new element of 
difficulty in the shape of increased expense. Simplicity of operation 


has been obtained by complication of the parts of the machine 
so that such instruments are much more expensive than 
those usually employed by skilled electricians. The simple and 
inexpensive nature of the telephone, on the other hand, renders 
it possible to connect every man's house, office, or manufactory 
with a central station, so as to give him the benefit of direct 
telephonic communication with his neighbours, at a cost not greater 
than that incurred by gas or water. 

At the present time we have a perfect network of gas-pipes 
and water-pipes throughout our large cities. We have main pipes 
laid under the streets communicating by side pipes with the various 
dwellings, enabling the members to draw their supplies of gas and 
water from a common source. 

In a similar manner, it is conceivable that cables of telephone 
wires could be laid underground, or suspended overhead, com- 
municating by branch wires with private dwellings, country houses, 
shops, manufactories, &c., &c., uniting them through the main 
cable with a central office where the wire could be connected as 
desired, establishing direct communication between any two places 
in the city. Such a plan as this, though impracticable at the 
present moment, will, I firmly believe, be the outcome of the intro- 
duction of the telephone to the public. Not only so, but I believe 
in the future wires will unite the head offices of the Telephone 
Company in different cities, and a man in one part of the country 
may communicate by word of mouth with another in a different 

I am aware that such ideas may appear to you Utopian and 
out of place, for we are met together for the purpose of discussing 
not the future of the telephone, but its present. 

Believing however as I do that such a scheme will be the ultimate 
result of the telephone to the public, I will impress upon you all 
the advisability of keeping this end in view, that all present arrange- 
ments of the telephone may be eventually realised in this grand 

The plan usually presented in regard to private telegraphs is 
to lease such lines to private individuals, or to companies at a fixed 
annual rental. This plan should be adopted by you, but instead 
of erecting a line directly from the one to another, I would advise 
you to bring the wires from the two points to the office of the 
Company and there connect them together ; if this plan be followed 
a large number of wires would soon be centred in the telephone 
offices, where they would be easily accessible for testing purposes. 
In places remote from the office of the Company, simple testing 
boxes could be erected for the telephone wires of that neighbour- 
hood, and these testing places could at any time be converted into 
central offices when the lessees of the telephone wires desire inter- 

In regard to other present uses for the telephone, the instrument 
can be supplied so cheaply as to compete on favourable terms with 


speaking tubes, bells, and annunciators, as a means of communica- 
tion between different parts of the house. This seems to be a very 
favourable application of the telephone, not only on account of 
the large number of telephones that would be wanted, but because 
it would lead eventually to the plan of intercommunication referred 
to above ; I would therefore recommend that special arrangements 
should be made for the introduction of the telephone into hotels 
and private buildings in place of the speaking tubes and annunciators 
at present employed. Telephones sold for this purpose should be 
stamped or numbered in such a way as to distinguish them from 
those employed for business purposes, and an agreement should 
be signed by the purchaser that the telephones should become 
forfeited to the Company if used for other purposes than those 
specified by the agreement. 

It is probable that such a use of the telephone would speedily 
become popular, and that as the public became accustomed to the 
telephone in their houses, they would recognise the advantage of a 
system of intercommunication. When this time arrives, I would 
advise the company to place telephones free of charge for a specified 
period in a few of the principal shops so as to offer to those house- 
holders who work with the central office, the additional advantages 
of oral communication with their tradespeople. The central office 
system once inaugurated in this manner would inevitably grow to 
enormous proportions, for those shopkeepers would thus be induced 
to employ the telephone, and as such connections with the central 
office increased in number, so would the advantages to house- 
holders become more apparent and the number of subscribers 

Should this plan be adopted, the company should employ a man 
in each central office for the purpose of connecting wires as desired. 
A fixed annual rental could be charged for the use of wires, or a 
toll could be levied. As all connections would necessarily be made 
at the central office, it would be easy to note the time during which 
any wires were connected and to make a charge accordingly 
bills could be sent in periodically. However small the rate of 
charges might be, the revenue would probably be something 

In conclusion, I would say that it seems to me that the telephone 
should immediately be brought prominently before the public, 
as a means of communication between bankers, merchants, manu- 
facturers, wholesale and retail dealers, dock companies, water 
companies, police offices, fire stations, newspaper offices, hospitals 
and public buildings, and for use in railway offices, in mines and 
[diving] operations. 

Agreements should also be speedily concluded for the use of 
the telephone in the Army and Navy and by the Postal Telegraph 

Although there is a great field for the telephone in the immediate 
present, I believe there is still greater in the future. 


By bearing in mind the great object to be ultimately achieved, 
I believe that the Telephone Company can not only secure for 
itself a business of the most remunerative kind, but also benefit the 
public in a way that has never previously been attempted. 
I am, gentlemen, your obedient servant, 


The American District Telegraph system differed from the London 
District system described on p. 80. It was established in 1870, and 
was very flourishing at the time of the introduction of the telephone. 
According to Edward A. Calahan l it owed its origin to the per- 
petration of a series of burglaries at Englewood, New Jersey, 
including the house of E. W. Andrews, first president of the Gold and 
Stock Telegraph Co., who had resigned his office in order to introduce 
the ' stock ticker ' into London. In his absence Mr. Calahan 
devised a system, and on Mr. Andrews' return the American District 
Telegraph Co. was formed. Work was commenced at Brooklyn 
Heights, and 100 subscribers were secured within a week. Whilst 
still retaining the protective features of ' police,' ' fire,' and similar 
calls, the principal feature of the Company's operations was to 
provide messengers on a request being conveyed telegraphically. 
The messengers were held in readiness at a central office serving 
a particular district. Lines radiated from this central office, 
and each line was ' looped in ' at the offices of several subscribers. 
A call box was placed in each subscriber's premises. To 
call the central office the subscriber pulled down a lever 
which was held up by a spring. In course of the return of the 
lever to its normal position a series of electric impulses were 
sent over the line to the central office, the impulses varying with 
each subscriber according to the arrangement of the call box. 
These impulses were received at the central office and recorded 
on a Morse register. The attendant there reading the number of 
the subscriber dispatched a messenger to the address indicated. 
The messenger was then at the service of the subscriber to carry a 
message, letter, or parcel, as might be required, payment being 
made according to the time occupied. 

The circuit and apparatus of the District Telegraph system 
may be observed from fig. 26, p. 97, by disregarding the telephone 
apparatus included in that illustration. 

The telephone was probably first applied as an auxiliary to the 
district call box to communicate only with the central office. 
' Instructions to Agents,' dated November 15, 1877, signed by 
Hubbard and Sanders, provide that a discount of twenty per cent, 
may be made ' For District Telegraph purposes.' Thomas B. 

1 Electrical World, N.Y., March 16, 1901, xxxvii. | 8. 


Doolittle, who filed an application for a patent on April 10, 1878, 
refers to the ' District Telephone System,' and says : 

In the use of the telephone system as it exists at present, where 
more than two stations are connected with a line wire, there is no 
practical means, that I am aware of, of preventing a message between 
any two stations from being heard or picked up at any or ah 1 of the 
other stations. This condition of things is a great disadvantage in the 
use of the telephone in a district system, where notice is sent to a 
main office to respond to an inquiry, or a command given to put 
two stations in communication. 1 

Mr. Doolittle's invention contemplates cutting out intermediate 
telephones on the same line, showing that it was customary to 
add telephones in the manner adopted for district call boxes, a 
number being in series on a single circuit, and it would imply that 
circuits were coupled in response to a demand to put two sub- 
scribers in communication. 

A development from this condition was the -addition of a direct 
line to which the telephone was attached. Mr. C. E. Scribner 
states : 

In Chicago the first exchange was established and the central office 
equipment placed in a small room in the back of an American 
District Telegraph Office in La Salle Street. The few subscribers 
to the system had the lines over which they telephoned connected 
with a small switchboard, and a single operator responded to their 
calls and made the connections they required. Each subscriber 
had, in addition to his own telephone line, a connection with the 
American District Telegraph system, and was provided at his office 
with the ordinary district call box, by means of which he transmitted 
a signal to indicate his wants. A register at the central office 
recorded telegraphically upon a tape the signals as received, and it 
was in response to these recorded signals that a connection with the 
subscriber's individual telephone line was made. 2 

Mr. Scribner was giving evidence at first hand but after a con- 
siderable interval of time. A contemporary record is to be found 
in an article by Mr. Haskins in ' La Lumiere Electrique ' of 1880, 
of which the following is a translation in part : 

There exist in America several systems for the establishment of 
telephone communications between private people, and the most 
important of these is the one known under the name of ' American 

1 U.S. specification, No. 209,115, October 22, 1878. 

2 Western Electric Company v. Capital Telephone and Telegraph Com- 
pany. Circuit Court of the United States, 1896. Supplemental Brief for 
Complainant, p. 18. 


district system/ which has, until now, been employed by the 
Edison-Gray Company. The number of telephones in use in this 
Company's service is equal, to-day, to approximately 1600, and 
the subscribers' list increases every day. 1 The total number of 
connections established every day amounts to several thousands, 
and our lines are capable of giving telephonic communications at 
a distance of seven and even nine miles. 

Signal lines are wires which pass through district boxes and 
end at the subscribers' residences or offices in order to enable them 
to ask either for messenger boy, a telephone, a cab, a policeman, 
or for a fireman. The circuit of this wire is closed or ' metallic,' 
and the signals are obtained, as in the systems of the same kind, 
by the cutting-out of the current. It is only after the wire has 
gone through a relay that it goes into the central exchange, this 
relay actuating a second circuit in which is interposed a Morse 

The district box is composed of a cylindrical case bearing a dial 
upon which moves an index, and the portion of the circumference 
through which this index is moved shows division marks correspond- 
ing to the different kinds of service orders which may be given. 
At the back part of the case is adapted a pawl or lever retained 
by the teeth of a ratchet wheel which permits, when pulled 
from the upper part towards its lower part, to wind up the spring 
of a clock mechanism disposed in such a manner that, through 
the register, it provides for closings of the current in a number 
more or less great according to the part of the arc followed by 
the index, and which may thus determine the inscription of the 
signal. In order to obtain a given signal it is sufficient to push 
down the level until the index comes in front of the order which 
it is desired to be transmitted. After releasing the lever the 
clock mechanism, while moving, will determine 3, 4, 5 or 6 
closings of the current, according to the order to be transmitted. 
This mechanism could even be disposed in such a way as to 
simplify the combinations in order to lengthen or to shorten the 
closings of the current. 

All the wires are run into the operators' room, each wire 
ending at a small brass plate i inch long, \ inch wide, and inch 

Each brass strip is provided with two holes disposed to receive 
plugs. Each one of the boards is provided with a telephone having 
one end of its wire connected to ground and the other end connected 
with a flexible conductor ending in a brass plug adapted to fit the 
holes of the brass strips. Let us assume now that Mr. A. has sent 
a signal to the central exchange, this signal having the number 26 
and corresponding to ' telephone.' The operator in charge of the 
signals immediately consults his register book and finds that the 

1 There are in Chicago two telephone companies : the Bell Company 
and the Edison-Gray Company. More than 3000 subscribers are divided 
about equally among these two companies. [Note in original.] 



signal 26 which has been sent corresponds to the board No. 984. 
He takes a ticket printed as follows : 



...... 984 



Underneath the word connect he writes the number of the 
calling subscriber, i.e. No. 984, and before this number he reproduces 
the signal which has been sent. This ticket is given to the operator 
of the nearest board, who immediately inserts the plug of its 
telephone flexible conductor in the plate No. 984 ; then he depresses 
the key of his manipulator in order to connect the battery to the 
line, and he advises Mr. A. that his signal has been received. Then 
Mr. A. says that he desires to speak to Mr. C. The number of 
Mr. C. is 516 ; then the operator writes down his signature on the 
ticket below the signature of the first operator, and he writes the 
No. 516 below the No. 984, and at the same time he writes down 
the hour. The ticket is then given to another operator whose 
apparatus has ordinarily no connection to the ground, but which 
possesses two keys, one for the left, one for the right. These keys 
have their connections established in the following manner : The 
person who calls being always at the right, the operator speaks 
through a tube to the operator of the main switchboard and asks 
for the numbers 984 and 516. The wires which establish the 
communication between the telephone of these numbers and the 
main switchboard are connected to the two plates which correspond 
to them, and the operator depresses the left key of the manipulator 
and listens, while effecting at the same time two or three nota- 
tions. In consequence of the depression of the key, the left wire 
is put in connection with the telephone and the battery is brought 
into circuit. At the moment he has received the answer of the 
left correspondent, he operates the right key and calls the other 
correspondent ; after this, the two correspondents may speak to 
one another. Generally the operator does not speak, he only 
listens for a sufficient time to ascertain that the two called corre- 
spondents are actually talking together, and when he is sure of this 
he calls the operator of the main switchboard to ask him to connect 
together the two correspondents and to disconnect his apparatus 
from the circuit in order to be free to establish new communications. 

When the correspondents have terminated their conversation 
they both place the index of their boxes in front of the signal No. 2, 
which means ' clear out the connection,' and the wires are 
immediately separated from one another. It takes much longer 


to describe all these operations than to do them, and yet, however 
rapid the system is, it is far from being quick enough on the days 
when there are no end of dispatches to be exchanged. 1 

In their paper read before the American Electrical Society in 
December 1879, Haskins and Wilson regarded the American District 
Telegraph system as then 

too well known to require a detailed description. It is only neces- 
sary to state briefly, that in Chicago the larger portion of all the 
calls from telephone subscribers are received at the central office 
by that system over a separate wire. . . . For this purpose the 
American District Telegraph Co. of Chicago have used their already 
established wires with the apparatus formerly employed exclusively 
for A.D.T. calls. ... To adapt this apparatus to the telephone 
exchange it was only necessary to add two signals to those already 
in use ' Telephone use ' and ' Telephone through/ the former 
signifying that the central office is wanted, the latter that the 
subscriber desires to be disconnected. 2 

Figure 26, which illustrates the circuit and apparatus of the 
American District Telegraph system and of the telephones used 
therewith, is taken from Firman's U.S. patent No. 328,305, dated 
October 13, 1885 (application filed January 16, 1880). 

Whilst the telephone exchange was suggested by Bell and his 
associates, it apparently developed in practice in various places by 
means of the American District Telegraph Service, and the application 
of telephones to the Law Telegraph system, as well as from the 
pioneer work on Holmes's Boston Burglar Alarm service and the 
Hartford Drug Store connections of Isaac D. Smith. The tele- 
graphic exchanges such as have been described as existing in New- 
castle, Philadelphia, New York, and elsewhere doubtless had some 
influence, but Dumont's proposals were apparently forgotten, his 
patent being only brought to light by research due to litigation 
in after years. 

Thus during Bell's absence from the United States the exchange 
system was rapidly progressing, the existence of numerous district 
telegraph systems facilitating this progress. The first fully equipped 
commercial telephone exchange established for public or general 
service was opened for business on January 28, 1878, at New Haven, 
Connecticut. This example was speedily followed by other cities, 
such as Bridgeport, New York, and Philadelphia, until it became 
evident to others than the Bell associates that the telephone was an 
invention to be reckoned with. 

The Western Union Telegraph Company was at that time the 

1 La Lumiere Eleclrique, ii. 155. 

2 Journal of the American Electrical Society, 1880, pp. 50-51. 



Station 3. 

FIG. 26. Telephone Exchange System as applied to American district 
telegraph by Firman. 


largest electrical organisation in the United States. Besides 
conducting ordinary telegraph business, it controlled the Gold and 
Stock service, and supplied customers with private lines to which 
were attached printing and dial telegraphs. 

The Bell telephone patents were offered to the Western Union, 
but they were declined. When some of their printing telegraph 
instruments were being taken out and superseded by Bell tele- 
phones, the \Vestern Union Company promptly realised that it was 
desirable to add the telephone to their telegraph business, and they 
did so, not in accord with, but in opposition to, the Bell interests. 

The Western Union was better equipped than any other organisa- 
tion in the United States to handle such a business. Their agents 
were everywhere, and their construction staffs were numerous and 
experienced. They needed only telephones and names to sail 
under. Edison had for years past been an adviser of the Western 
Union ; Gray had been indirectly connected with them ; Dolbear 
had claimed that the permanent magnet telephone was his invention : 
so the Western Union announced its entry into the telephone field 
under the auspices of these three. The first had won his spurs in 
telegraphic inventions which the Western Union had acquired ; 
the second had deposited a ' caveat ' for a telephone on the same 
day as Bell had filed his specification ; and the third was not un- 
known in scientific and technical circles. Of the three the only 
one to make an important contribution to the industry was the 
first mentioned. Apart from the subsequent litigation, and the 
commercial development arising from the competition, which will 
later be referred to, the entry of the Western Union Company 
into the telephone field is chiefly remarkable for the introduction of 
the battery transmitter. 



IN describing the apparatus exhibited at the Centennial Exhibition 
at Philadelphia in Chapter VI, the transmitter was reserved for 

FIG. 27. Bell's Centennial Liquid 
Transmitter (perspective) . 

FIG. 28. Bell's Centennial 
Liquid Transmitter (section) . 

later reference. The liquid transmitter there exhibited is illustrated 
in perspective by fig. 27 and in section by fig. 28. It is thus 
described by Bell : 

D is the diaphragm ; R is the rod attached to the centre of the 
diaphragm, and dipping into the liquid in the cup C. F, supporting 
frame, and M the mouth-piece. This also has a straining ring with 
adjusting screws like that used in the single and double pole mem- 
brane telephones already described. Indeed, they were all three 

99 H 2 


made from the same casting. The cup is supported upon a screw, 
by which it can be adjusted vertically to vary the depth of immer- 
sion of the lower end of the rod R. 1 

It will be seen that this Centennial liquid transmitter illus- 
trates in a practical form one of the methods other than inductive 
action of obtaining undulations in the circuit suggested by Bell 
in his patent, from which the following is quoted : 

Electrical undulations may also be caused by alternately increasing 
and diminishing the resistance of the circuit or by alternately increas- 
ing and diminishing the power of the battery. The internal resist- 
ance of a battery is diminished by bringing the voltaic elements 
nearer together, and increased by placing them farther apart. The 
reciprocal vibration of the elements of a battery, therefore, occasions 
an undulatory action of a voltaic current. The external resistance 
may also be varied. For instance, let mercury or some other 
liquid form part of a voltaic circuit, then the more deeply the con- 
ducting wire is immersed in the mercury or other liquid, the less 
resistance does the liquid offer to the passage of the current. Hence, 
the vibration of the conducting wire in mercury or other liquid 
included in the circuit occasions undulations in the current. 2 

Bell has explained that 

all the parts of the specification alluding to the variable resistance 
mode of producing electrical undulations were put in by [him] 
at the very last moment, before sending the specification off to 
Washington to be engrossed. 8 

It is for this reason probably that no drawing of the variable resist- 
ance method appeared in the specification, though Bell further 
explains that 

all that was necessary, it seemed to me, to illustrate this mode of 
producing electrical undulations was to attach the wire I desired 
to vibrate to the various instrumentalities shown and described in 
my specification and illustrated in figs. 5 and 7 [thereof]. That is, the 
water would take the place of the electro-magnet in the transmitting 
instrument, and if it was desired to produce undulations for the 
purpose of multiple telegraphy, the wire that dipped into the water 
would be attached to the free end of the reed illustrated in fig. 5 ; 
and if it was desired to transmit vocal or other sounds telegraphically 
including articulate speech the wire to be vibrated should be 
attached to the centre of the stretched m inbrane shown in fig. 7, 

1 Deposition, p. 99. 

* U.S. specification, No. 174,465, March 7, 1876 (application filed February 
14, 1876). * Deposition, p. 88. 


which would then have to be substantially horizontal to admit of 
the vertical vibration of the wire. 1 

The American Academy paper was presented May 10, 1876, 
between the date of the patent and the opening of the Centennial 
Exhibition. In this paper Bell said : 

Electrical undulations can be produced directly in the voltaic 
current by vibrating the conducting wire in a liquid of high resist- 
ance included in the circuit. ... A platinum wire attached to a 
stretched membrane completed a voltaic circuit by dipping into 
water. Upon speaking to the membrane articulate sounds pro- 
ceeded from the telephone in the distant room. The sounds 
produced by the telephone became louder when dilute sulphuric 
acid, or a saturated solution of salt, was substituted for the water. 
Audible effects were also produced by the vibration of plumbago in 
mercury, in a solution of bichromate of potash, in salt and water, 
in dilute sulphuric acid, and in pure water. 2 

The liquid transmitter was developed from a device designed 
by Bell some time previously to obviate the difficulties from induced 
currents in telegraphic instruments. The object and the means 
are thus explained in a specification which was prepared but not 
proceeded with : 

When a Morse Sounder or other telegraphic or electrical apparatus 
is placed in a voltaic circuit, the induction of the current upon itself 
in the coils of the electro-magnet or other instrument gives rise to 
an induced current of high tension. 

When the circuit is broken at any point, as by the lifting of the 
key, the induced current leaps across the break in the circuit in the 
form of a bright spark. When a powerful battery is employed, or 
when the circuit is very rapidly made and broken, the spark becomes 
so intense as to burn or oxidise the points between which it appears, 
and thus prevent the effective working of the instruments upon the 

This defect has hitherto been overcome by the employment of 
condensers, by means of which the spark has been materially lessened 
or destroyed. 

Now, I have discovered that the same effect may be produced 
by introducing between the points where the circuit is broken an 
imperfect conductor of electricity, which shall offer a very great 
resistance to the voltaic current, but afford a free passage for the 
induced current which occasions the spark. 

Such a substance is water, especially when slightly acidulated. 

1 Deposition, p. 87. Fig. 5 of the patent is illustrated in fig. 15, p. 58, 
and fig. 7 of the patent in fig. 4, p. 42. 

* Proceedings of the American Academy of Arts and Sciences, vol. xii. 
Researches in Telephony,' par. 13. 


Retort carbon, animal and vegetable tissues, and other substances 
offering a high resistance answer the purpose ; but I prefer to employ 
a liquid (like water) which can be decomposed by the passage of the 
current. 1 

\Yhen including the variable resistance method in his telephone 
patent it occurred to him that water was not a good illustrative 
substance to be specified in this connection on account of this very 
fact of its decomposability by the action of the current. He 
therefore proposed to use as a typical example a liquid that could 
not be thus electrolytically decomposed, and specified mercury as 
the best example of such a liquid known to him. 2 

The caveat deposited in the Patent Office at Washington by 
Elisha Gray on the same day as, though several hours later than, the 
specification of Bell also contemplated the use of a liquid transmitter. 
The caveat is quoted in ' The Speaking Telephone ' (Prescott), 
1878, p. 202, accompanied by an illustration. After a general 
description of the apparatus Gray says : 

Owing to this construction, the resistance varies constantly in 
response to the vibrations of the diaphragm, which, although irregu- 
lar, not only in their amplitude but in rapidity, are nevertheless 
transmitted through a single rod, which could not be done with a 
positive make and break of the circuit employed, or where contact 
points are used. 

I contemplate, however, the use of a series of diaphragms in a 
common vocalising chamber, each diaphragm carrying an indepen- 
dent rod, and responding to a vibration of different rapidity and 
intensity, in which case contact points mounted on other diaphragms 
may be employed. 

The vibrations thus imparted are transmitted through an electric 
circuit to the receiving station, in which circuit is included an 
electro-magnet of ordinary construction, acting upon a diaphragm 
to which is attached a piece of soft iron, and which diaphragm is 
stretched across a receiving vocalising chamber, A. 

The diaphragm at the receiving end of the line is thus thrown 
into vibrations corresponding with those at the transmitting end, 
and audible sounds or words are produced. 3 

It is not surprising that in an electrical atmosphere of such high 
tension as may be produced by strenuous commercial and legal 
contentions charges of plagiarism should have been made. Let it 
be understood that neither Bell nor Gray had at the time of filing 
their respective specification and caveat put a liquid transmitter 
into operation. Nor had Gray constructed or designed an electro- 

1 Deposition, p. 85. * Ibid. p. 87. 

3 The Speaking Telephone, Prescott, 1878, p. 204. 


magnetic receiver such as he outlined. Bell had transmitted and 
received vocal sounds through his magneto instruments, and had 
thereby been satisfied that his conception of the undulatory current 
theory was right. His aim was to protect without delay the 
undulatory current and the means for obtaining that current. One 
means was inductive action, another was variable resistance. The 
liquid transmitter was an illustrative instrumentality of the latter 

Gray's proposition to elaborate his transmitter by increasing 
the number of diaphragms, each responding to a different note, is 
an indication that he had not reached the stage which Bell had in 
recognising that the single diaphragm would serve to translate the 
aerial vibrations into corresponding electrical currents. The idea 
in Gray's mind was apparently analogous to the harp telephone 
suggested, but not constructed, by Bell in the early stages of his 
investigations. Such an idea was but a stepping stone to Bell. 
By its aid and that of other stepping stones already detailed he 
reached, as his patent disclosed, the simple method by which a 
single diaphragm transmitted not merely a simple vibration but 
the resultant of complex vibrations. 

A caveat is not intended to be a finished or complete specifica- 
tion. Its having been deposited is proof of date, and that is its 
main purpose. Gray's document was a caveat, the document 
which Bell had deposited some hours earlier was a complete 

But Gray's caveat was accompanied by a drawing of his liquid 
transmitter and an illustration of an individual talking into it. 
Bell's liquid transmitter was only described, and there was no such 
pictorial representation of talking people as Gray depicted. These 
were no doubt contributory causes to the claims set up for Gray 
and for the persistence with which his friends maintained that he 
invented the telephone. The Western Union Company had aban- 
doned Gray's claims in 1879, but they were revived by others in 
later suits. 

Gray's work in this field was well known to his telegraphic 
contemporaries, whilst Bell was comparatively unknown. It is not 
surprising that the surface indications of Gray's caveat, together 
with his recognised standing in the electrical world, should have 
produced an impression that served to influence the judgment of 
his friends. But going beneath the surface and examining the 
specification of Bell and the caveat of Gray, it is easy to see that Gray 
showed the externals without the essentials, and Bell exactly and 
specifically defined the essentials, but was content with a very 
general indication of some of the means. 

Gray's words indicate that makes and breaks will not produce 


the required result, but that a single rod will permit the transmission 
of the vibrations of the diaphragm, which is suggestive of a recog- 
nition of the need of an undulatory current. But the practical 
feature of this suggestion is destroyed when he goes on to contemplate 
the use of several diaphragms, which were apparently intended to 
respond to different vibrations, and by co-operation were to send 
to the line a composite result. 

Bell, on the contrary, emphasised the fact that in order to trans- 
mit speech it was necessary to put on to the line a current of a 
character that had not hitherto been used. He denned those which 
had been used and the kind of current that needed to be employed. 
He said in effect : You must have an undulatory current. If you 






FIG. 29. Gray's Liquid Transmitter. 

talk to a diaphragm it will be set in motion by the sound waves, 
which are undulating. The undulatory current was the central 
and essential feature of Bell's patent, together, of course, with the 
means for obtaining it. These means were in effect three. The one 
which was illustrated, and therefore received most attention, was 
the use of a diaphragm as an electric generator, which by inductive 
action should produce a current varying in accordance with the 
vibrations imparted to the diaphragm by the air waves. 

But, in effect said Bell, this absolutely essential undulating 
current may also be obtained by using the vibrations of the 
diaphragm to control a current otherwise obtained. You may go 
to the source and vary the position of the elements of a cell, or 
you may interpose an instrument in the external circuit which shall 
vary the current in accordance with the motion of the diaphragm. 



The inclusion of such a bold feature as the variation of the 
source is an indication of Bell's intention to control all available 
methods, and a further proof of his perception of the essential 
feature for the transmission of speech. With this in his mind it 
became unnecessary to define with too great exactness, or even to 
illustrate by a drawing, any particular means of varying the external 
resistance. He contented himself with the statement of the fact 
that it can be so varied, and described such a method as his ' mercury 
or other liquid ' transmitter with a preliminary ' for instance.' 

Notwithstanding the inclusion of this description in Bell's 
patent, it was even suggested in court that his Centennial liquid 
transmitter was copied from that shown in Gray's caveat. Such 
a suggestion was indignantly combated by Bell's counsel ; but it 
is only necessary to examine both designs to see how groundless 
such suggestions were. Fig. 29 M 

represents Gray's Liquid Trans- 
mitter and fig. 30 that of Bell. 

Gray's description is as 
follows : 


FIG. 30. Bell's Liquid 

A box or chamber, A, across 
the outer end of which is 
stretched a diaphragm, ... of 
some thin substance, such as 
parchment or gold beaters' skin, 
capable of responding to all 
the vibrations of the human 
voice, whether simple or com- 
plex. Attached to this dia- 
phragm is a light metal rod, 

[a], or other suitable conductor of electricity, which extends into 
a vessel, B, made of glass or other insulating material, having 
its lower end closed by a plug, which may be of metal, or through 
which passes a conductor, . . . forming part of the circuit. 

This vessel is filled with some liquid possessing high resistance, 
such, for instance, as water, so that the vibrations of the plunger 
or rod, [a], which does not quite touch the conductor, . . . will cause 
variations in resistance, and, consequently, in the potential of the 
current passing through rod [a]. 1 

Bell's description is as follows : 

D is the diaphragm, R is the rod attached to the centre of the 
diaphragm, and dipping into the liquid of the cup C. F, supporting 
frame, and M the mouthpiece. . . . The cup is supported upon 

1 The Speaking Telephone, Prescott, 1878, pp. 204, 216. 


a screw, by which it can be adjusted vertically to vary the depth 
of immersion of the lower end of the rod R. 1 

While necessarily very similar in appearance, there is an obvious 
difference in design and operation. Gray's vessel was to be of glass 
or other insulating substance. The liquid was to have a high 
resistance, and its function was to complete the circuit between the 
point of the rod vibrated by the diaphragm and the plug or point 
at the base of the vessel. The resistance varied with the distance 
between the point of the rod and the metal plug or conductor at 
the base of the vessel. 

In Bell's design the variations in the current were obtained by 
the greater . or lesser surface contact of the rod, according to the 
extent of its immersion in the liquid. 

This Centennial liquid transmitter of Bell's simply put into a 
practical form the method of varying the external resistance which 
was described in his original patent. The use of a liquid for such a 
purpose seems to have occurred to ah 1 who devoted their attention 
to the subject, including Edison, who also in his early experiments 
was following the same line as Gray in seeking utility from multiple 
diaphragms. This feature serves to distinguish Bell from all his 
competitors, and indicates that they had no conception that the 
translation from sound to current could be so simply attained. And 
the reason undoubtedly is that they had not mastered, as Beh 1 had, 
the acoustic fact and its electrical equivalent. As Professor Clerk 
Maxwell said in his Rede lecture delivered at Cambridge in May 
1878, Professor Graham Bell, the inventor of the telephone, was ' not 
an electrician who had found out how to make a tin plate speak, but 
a speaker who, to gain his private ends, had become an electrician.' 2 

It will be seen how far behind were those experimenters who 
sought in multiple diaphragms some aid in translating compound 
sounds into electrical currents. This is the more surprising when 
Reis was content with one and when the phonautograph of Leon 
Scott gave graphic representations of speech waves with but a single 
diaphragm. In truth, Bell's contemporaries were not only far behind 
Bell but in this regard also behind Reis, an account of whose work 
is, for greater convenience, reserved for a later chapter. Gray had 
illustrated a single diaphragm but was evidently not confident 
that it would suffice. A similar lack of confidence was shown by 
Edison in his early efforts. But the first commercial forms of 
variable resistance transmitter were contributed not by either 
Bell or Gray, but by Edison in the early stages, and by modifications 
of the discoveries of Hughes later. 

1 Deposition, p. 99. 

5 The Scientific Papers of James Clerk Maxwell, ii. 742. Nature, vol. xviii. 


As we have already seen, when the Western Union Telegraph 
Company decided to go into the telephone business, Bell's magneto 
telephone was being used extensively for private line purposes. It 
was this practical development which sufficiently demonstrated that 
the telephone was no toy, but 'a business proposition.' Equally 
clear would it be to them that the feeble results should be capable 
of improvement. The very simplicity of the magneto telephone, 
the independence of battery attained by the use of a permanent 
magnet, were in the nature of temptations to continue the inductive 
method. The very fact that telephones talked, and were intended 
to be placed in the hands of the general public and not of expert 
electricians, would tend to justify the retention of simple forms. 
Only with the growth in its use, only when greater demands were 
made upon it, did the limitations of the magneto instrument as a 
transmitter become so apparent as to lead to an organised attempt 
at improvement. And during this eventful period, of its history 
the Western Union Telegraph Company decided to become interested 
in it, and Bell himself was abroad. 

Thomas Alva Edison was an expert upon whom the Western 
Union were justified in placing reliance. In various branches 
of telegraphic work he had proved his ability. He was commis- 
sioned to make improvements in the telephone, and he did. 

His achievements in the telephone art are thus recounted by 
Messrs. Dyer and Martin : 

With his carbon transmitter he gave the valuable principle of 
varying the resistance of the transmitting circuit with changes in 
the pressure, as well as the vital practice of using the induction coil 
as a means of increasing the effective length of the talking circuit. 1 

In the application of the induction coil or transformer in con- 
nection with speaking telephones Edison was original, 2 but in the 
application of the variable resistance he introduced no new principle, 
but simply ascertained by experiment a suitable method of applying 
the principle fully described by Bell in his first patent. 

In the United States patent system applications for patents 
which show the same object are ' put into interference ' with a view 

1 Edison. His Life and Inventions, p. 180, by Frank Lewis Dyer and 
Thomas Commerford Martin. Harper & Bros. Mcmx. 

2 An induction coil was applied by Dr. Wright to the Reis transmitter, 
a condenser being used as a receiver : ' The line current is made to pass 
through the primary of a small induction coil. In the secondary circuit he 
places two sheets of paper, silvered on one side, back to back so as to act as 
a condenser. Each current that comes from the sending apparatus produces 
a current in the secondary circuit, which charges and discharges the condenser, 
each discharge being accompanied by a sound like the sharp tap of a small 
hammer. The musical notes are rendered by these electric discharges and are 
loud enough to be heard in a large hall.' Electricity, Ferguson, 1866, p. 258. 


to determine by evidence and examination the priority of invention. 
The claims of Bell, Gray, Edison, Dolbear and others were ' in 
interference/ and were adjudicated upon by ' the Examiner of 
Interferences,' who reported on July 21, 1883. The following are 
extracts from his report : 

The first conversation said to have taken place between Orton 
and Edison relative to the subject of transmitting speech was in 
July 1875, but the first conversation relative to transmitters is 
not said to have taken place until in March or April 1876. In 
November or December 1875, Edison first heard of Bell as a person 
working in the same line as Gray on harmonic telegraphy, hence 
in the same line as he, Edison, had been engaged by Orton to work 
up for the Western Union. Bell's patent of March 7, 1876, was 
undoubtedly granted prior to the interview with Orton in which 
the transmitter is said to have been first mentioned, and it is but 
fair to assume, though the record is silent on the point, that Edison 
was at the time familiar with Bell's specification, in which various 
modes of producing undulatory currents are mentioned. Following 
closely on the grant of this patent, and contemporaneous with the 
construction of the alleged first successful transmitter for articulate 
speech, was the description given by Johnson to Edison of Bell's 
Centennial exhibition of the telephone. This exhibition took 
place June 25, 1876, and was not only preceded by Bell's patent 
of March 7, 1876, but also by his lecture read before the American 
Society of Arts and Sciences, May 10, 1876, and published shortly 

A knowledge of these transactions and publications is not 
brought home to Edison, for he is not interrogated with respect 
thereto ; but being public, and Edison interested as he was in 
matters pertaining to electricity, more especially that branch upon 
which he knew Bell was engaged, it may be reasonably inferred 
from the coincidence in time and other circumstances that if not 
already fully informed as to Bell's claims and inventions, as seems 
most probable, he began at once to investigate the subject when 
he learned from Johnson that articulate speech had actually been 
transmitted and reproduced. Unquestionably Edison was well 
acquainted with the abstract principle that a rise and fall in tension 
could be produced by varying the resistance, as indicated in his 
previous experiments and inventions ; but when he heard of Bell's 
success at the Centennial, it would be only natural that he should 
look to Bell's prior patents for suggestions. 

It was in July 1876 that Edison learned of Bell's Centennial 
exhibition of the articulating telephone, and in the same month 
we find the first evidence of a distinct effort being made on the 
part of Edison to construct an instrument designed solely for the 
transmission of speech. . . . Commencing with July 1876, the record 
indicates the active prosecution of the subject of the speaking 
telephone. Two lines of experiments seem to have been followed. 


the first currently designated the ' water telephone,' wherein the 
tension of the current in a closed circuit was to be varied by the 
movements of electrodes with reference to an interposed liquid, and 
the second wherein the vibration of a diaphragm or diaphragms 
imparted motion to several contact points and cut in and out 

It will be found upon inspection of Bell's patent of March, and 
his lecture of May 1876, that the methods upon which work was 
actually begun and prosecuted under Edison's supervision are 
all suggested as equivalent or alternate modes of inducing electrical 
undulations, and it is a coincidence that these experiments should 
have begun so soon after public notice had been directed to Bell's 
patent by the exhibition of his telephone, provided Edison had 
received no information or suggestion therefrom. 1 

From the same source we learn : 

On the next day, March 22, 1877, an agreement was entered 
into between Edison and the Western Union Telegraph Company, 
whereby the former agreed to assign all the inventions and im- 
provements capable of being used on telegraph lines he should 
make during the period covered by the agreement, the latter to 
remain in force for five years unless sooner terminated, and to 
take effect March i, 1877. On March 23, 1877, the day following 
the signing of the agreement, directions were given for the prepara- 
tion of Case 130, the first application filed by Edison referring to 
the electrical transmission of articulate speech. 2 

As an expert of the Western Union Company Edison entered 
the telephone field. ' The work that Edison did was, as usual, 
marked by infinite variety of method as well as by the power to 
seize on the one needed element of practical success.' 3 

As we have seen, the principle upon which Edison worked had 
been defined previously by Bell. It was the principle of varying 
the resistance of the circuit in correspondence with the excursions 
of the diaphragm. It remained for Edison to select a ' tension 
regulator ' which should be effective and adapted to commercial 
use. A material subject to evaporation like water, as suggested 
by Gray and devised by Bell, was obviously ill adapted for com- 
mercial use. There was probably no one so well equipped as 
Edison to discover the best material and method for obtaining the 
required variations in resistance. Long experience in the tele- 
graphic field gave him the starting point of knowledge. His 
experimental laboratories and skilled assistants enabled him to 
proceed at once with likely experiments and carry them to a con- 
clusion. Perhaps also his characteristic method of reaching an 

1 The Speaking Telephone Interferences, pp. 180-2. * Ibid. p. 203. 

3 Edison, His Life and Inventions, p. 178. 


end by the process of exhaustion of means was equally helpful 
in this case. 

On July 20, 1877, he filed an application which is a distinctly 
bad start from the acoustic standpoint, since the apparatus 
consists of 

a resonant case with several tympans, adjusted to different degrees 
of tension or delicacy, and these are all so connected with contact 
points in the metallic line circuit that the electric pulsations will 
be sent over the line from one or more of these tympans, and 
operate upon an electro-magnet and receiving tympan of a resonator. 1 

It seems somewhat surprising that at this date Edison should have 
sought assistance from multiple diaphragms and ' resonant boxes/ 
but the specification is of interest from the casual mention that the 
circuit closing springs shall be ' preferably with carbon or plumbago 

It was in the direction of the use of carbon or plumbago that 
Edison was to contribute to the art. A few months later (December 
13, 1877) he applied for a patent z in which the variable resistance 
first described was the familiar rheostat composed of convolutions 
of wire capable of being more or less short circuited by a metallic 
contact operated by the diaphragm. But alternatives were 
suggested, and amongst them ' a semi-conductor such as plum- 
bago,' which, however, was intended to operate in the same way 
as the rheostat, for ' the forward movement of the diaphragm 
causes more and more platina to come in contact with the plumbago, 
thus allowing the greater part of the current to pass through the 
platina, according to the amplitude of the diaphragm vibrations.' 
This patent is also noticeable for its inclusion of the induction coil 
and in that a single diaphragm was shown. 

But the practical variable resistance had not yet been reached. 
The material was there but the disposition was unsuitable. Yet 
years before Edison had devised a rheostat following the lines 
of Clerac, in which carbon was inserted directly in the circuit. 
Harking back on this experience, Edison commenced to experiment 
on more definite lines. He sought to obtain variations in resistance 
by the operation of the diaphragm on some substance directly in 
the circuit. His carbon rheostat varied its resistance by pressure. 
A suitable material should respond to the action of the diaphragm 
and transmit more or less current, in accordance with the greater 
or less excursions of the diaphragm. Silk fibres (or ' fluff,' as 
it was described) covered with plumbago, and various other applica- 
tions of the same idea, finally led to the filing of an application on 

1 U.S. specification, No. 203,014, April 30, 1878. 
* U.S. specification, No. 203,013, April 30, 1878. 



March 7, 1878, in which it is explained that the carbon heretofore 
employed in connection with a diaphragm is of too great resistance 
to be adapted for use in the primary circuit of an induction coil. 
In this specification is also recorded the discovery that lamp black 
obtained from the combustion of very light hydrocarbons, such as 
gasoline or naphtha, can be used. 

I select from lamp black thus made only the very blackest 
portions and then place the same in a mold, and subject it to a 
very powerful pressure, sufficient to consolidate the same and place 
it in a correspondingly shaped [i.e. to the mould] cavity contiguous 
to the diaphragm, with a piece of cork or a piece of rubber 

FIG. 31. Edison Carbon 
Transmitter, as illustrated 
in patent. 

FIG. 32. Edison Car- 
bon Transmitter (Clarke's 

intervening between the same and the diaphragm, and connect the 
discs of platina foil that are used at each side of the carbon in 
the primary circuit of the induction coil, and obtain from the 
pressure resulting from the motion of the diaphragm the necessary 
rise and fall of tension without the great resistance heretofore 
inseparable from the carbon in said circuit. 1 

The illustrations to this specification from which fig. 31 is 
obtained show the influence of the commercial form of Bell's 
telephone since the carbon arrangement is mounted in an instrument 
having the external form of the Bell hand telephone (fig. 31). This 
was not well adapted to the purpose, and before the Edison trans- 
mitter was commercially issued it underwent some modifications 
in minor points of detail at the hands of George D. Clarke 2 (fig. 32) 

1 U.S. specification, No. 203,016, April 30, 1878. 

2 U.S. specification, No. 217,773, July 22, 1879 (app. filed Sept. 16, 1878). 


and George M. Phelps l (fig. 33), the illustrations to whose patents 
will serve to indicate the developments. 

The theory upon which Edison worked, as stated by himself, 
was that the electrical resistance of carbon became materially 
lessened under pressure, and that the variations in the pressure 
upon the carbon mass accounted for and produced the necessary 
variations in the current. The theory, as we shall see, was to be 
open to question, but of more importance sometimes than theory is 

FIG. 33. Edison Carbon Transmitter (Phelps's illustration). 

practice. In practice Edison had placed his patrons, the Western 
Union Company, in possession of an instrument which so much 
increased the power and extended the range of telephonic speech as 
to place them in a commanding position at this eventful period of 
telephone exchange development. 

The transmitter, operating in the manner which he had defined, 
was to have a short life, but the use of the induction coil was 
permanent, and carbon, though in a form differing from that which 
he had selected, was not to be removed from its pride of place. 

1 U.S. specification, No. 214,840, April 29, 1879 (app. filed Dec. 6. 1878). 



OF more than passing interest is the international character of the 
scientific discoveries leading up to the invention of the telephone. 
Faraday, Henry, Volta, Oersted, Ampere, and others all wrested 
secrets from nature contributing material with which to build ; 
Wheatstone, Helmholtz, Page, and Reis served, each in his way, to 
indicate some appropriate method of using those materials. But 
it is of special interest to note the conjunction of Great Britain and 
America in the nationality of the two men who are more parti- 
cularly associated with the scientific and practical development of 
the telephone. Bell was not a citizen of the United States when 
Sir William Thomson saw and heard the telephone at the Phila- 
delphia Exhibition. But the telephone was invented in Boston 
by a native of Edinburgh who was to become a citizen of the 
United States, and the microphone was discovered in London 
by David Hughes, who, though a Londoner born, had spent the 
greater part of his life in the United States. When he was seven 
years of age (1838) he was taken with the family to settle in Virginia. 
Of Welsh origin, he inherited the national and family taste for 
music, and he obtained the position of professor of music at the 
College of Bardstown, Kentucky, subsequently occupying the 
chair of Natural Philosophy in the same college. Here he set out 
to devise a telegraphic instrument which should print the received 
message in Roman characters. He succeeded, and obtained its 
adoption not only in America but also throughout Europe. In the 
evening of his life he continued to take an interest as a scientist 
in discoveries which he had previously pursued as an inventor. 

Hughes was not the first to note that a variation in the current 
followed from varying degrees in the intimacy of contact between 
two sections of an electrical conductor. 1 Berliner had not only 

1 A. M. Tanner in the Electrical Review, London, November 21, 1890, 
p. 612, recounts prior observers, Du Moncel (1856), Mousson (1861), Buff (1865), 

113 I 


noticed it, but had claimed its use in a telephone transmitter. But 
Hughes was the first to study the facts and explain the causes to 
the satisfaction of the scientific world. His investigations were com- 
municated to the Royal Society by Professor Huxley on May 8, 1878. 
The telephone, then so recently invented, made the investigations 
possible, and he pays his tribute thereto in his opening words : 

The introduction of the telephone has tended to develop our 
knowledge of acoustics with great rapidity. It offers to us an 
instrument of great delicacy for further research into the mysteries 
of acoustic phenomena. It detects the presence of currents of 
electricity that have hitherto only been suspected, and it shows 
variations in the strength of currents which no other instrument 
has ever indicated. 1 

Hughes's investigations were originally undertaken in order to 
observe the effect of sonorous vibrations upon the electrical be- 
haviour of matter. Sir William Thomson and others had shown 
that the resistance to the passage of currents offered by wires 
was affected by their being placed under strains, and Hughes 
believed that the wire would vary in its resistance when it was 
used to convey sound 2 so he commenced with a study of the 
influence on the current of a wire under stress, but found no change 
indicated in the telephone until the wire broke. He then sought 
to imitate the condition of the wire at the moment of rupture by 
replacing the broken ends and pressing them together with a 
constant and varying force by the application of weights. 

It was soon found that it was not at all necessary to join two 
wires endwise together to reproduce sound, but that any portion 
of an electric conductor would do so, even when fastened to aboard 
or to a table, and no matter how complicated the structure upon 
this board, or the materials used as a conductor, provided one or 
more portions of the electrical conductor were separated and only 
brought into contact by a slight but constant pressure. Thus, if the 
ends of the wire terminate in two common French nails laid side by 
side, and are separated from each other by a slight space, were electri- 
cally connected by laying a similar nail between them, sound could 

1 Proceedings of (he Royal Society, May 9, 1878, xxvii. 362. 

2 Bell had the same impression. In a letter to Mr. Hubbard dated 
May 4, 1875, he says : ' Another experiment has occurred to me which, 
if successful, will pave the way for still greater results than any yet obtained. 
The strings of a musical instrument in vibrating undergo great changes of 
molecular tension. In fact the vibration represents the struggle between 
the tension of the string and the moving force impressed upon it. I have 
read somewhere that the resistance offered by a wire to the passage of an 
electrical current is affected by the tension of the wire. If this is so, a con- 
tinuous current of electricity passed through a vibrating wire should meet with 
a varying resistance, and hence a pulsatory action should be induced in the 
current.' (Deposition, p. 53.) 


be reproduced. The effect was improved by building up the nails 
log-hut fashion into a square configuration, using ten or twenty 
nails. A piece of steel watch chain acted well. 1 

He relates the various substances which had been tried, and 
proceeds to describe the instrument which he called a microphone 
because it was suitable for magnifying weak sounds. The name was 
given by Wheatstone to a sort of duplex stethoscope, regarding 
which Wheatstone says : 

The microphone is calculated only for hearing sounds when it is 
in immediate contact with sonorous bodies : when they are diffused 
by transmission through the air, this instrument will not afford the 
slightest assistance. 2 

Although the members of the Royal Society attending the 
reading of Hughes's paper assumed that they heard the tramp 
of a fly upon the table, it must be recognised that what they heard 
was the sound created in the telephone by the mechanical dis- 
turbance of the conductor rather than the reproduction of any 
actual sound, however minute, made by the fly. The ' micro- 
phone ' must for our purposes be regarded as an effect described 
in the paper rather than as a magnifier of sound. It is quite evident, 
says Hughes 

that these effects are due to a difference of pressure at the different 
points of contact, and that they are dependent for the perfection 
of action upon the number of these points of contact. 3 

Hughes related the experiments which he had made in 
the transmission of vocal sounds on the lines of this discovery, 
and he called any of the preparations of finely divided metals or 
charcoal ' confined in a glass tube or a box and provided with wires 
for insertion in a circuit ' a transmitter. 

Edison's carbon telephone was called a transmitter and it was 
the most effective transmitter available. It is perhaps partly due 
to the use of this word by Hughes that Edison immediately launched 
an accusation of piracy against Hughes and Preece, to whose 
appreciation of the importance of the facts and kind counsel in 
the preparation of the paper the professor had expressed his 
indebtedness. The controversy waged furiously, but Sir William 
Thomson (Lord Kelvin) intervened, and in a public letter summed 
up the facts, concluding by saying : 

1 Proceedings of the Royal Society, May 9, 1878, xxvii. 362. 
1 Quarterly Journal of Science, 1827, Part II. Wheatstone's Scientific 
Papers, p. 33. 

3 Proceedings of the Royal Society, May 9, 1878, xxvii. 366. 

1 2 


I cannot but think that Mr. Edison will see that he has let 
himself be hurried into an injustice, and that he will therefore not 
rest until he retracts his accusations of bad faith publicly and 
amply as he made them. 1 

The following contemporary comment on the controversy 
represents the generally accepted view of the difference between 
the conceptions of Edison and of Hughes : 

The merit in Professor Hughes' discovery mainly consists in 
this that it is a property of the contact of two conductors, and 
that it is not confined to any one, but to all conductors. Edison's 
claim has been until now, and can only be stated as follows : 
' I discovered about two years ago that carbon of various forms, 
such as plumbago, graphite, gas retort, carbon, and lampblack, 
when moulded in buttons, decreased the resistance to the passage 
of the electrical current by pressure ' (we give his own words). 
This was an important discovery, but it had been anticipated by 
M. Clerac, of the French Telegraphic Administration, Paris, who, 
as long ago as 1866, constructed tubes containing powdered carbon, 
the electric resistance of which could be regulated by increasing or 
diminishing the pressure upon it by means of an adjusting screw. 
Mr. Edison applied it to his carbon telephone in a manner well known. 
Now it appears to us, from all accounts, that Mr. Edison, though 
very nearly discovering the full significance of the microphone, 
did not do so. He explained the real nature of the property which 
he had found to exist in carbon, and, let us say, the vague but 
all-comprehensive patent phrase, ' finely divided metals/ to be a 
diminution of resistance in the mass of the material. The discovery 
of the microphone limits it to an effect of the surfaces of contact ; 
but it is plain that Edison did not find this out, else we have no 
hesitation in saying so great an inventor would not have confined 
himself to buttons of carbon placed between platinum discs. Thus 
it is that Professor Hughes' discovery explains the true action of 
Edison's telephone, which is indeed a form, but a disguised form, 
of the microphone. It may enable the former to rid itself of its 
mask, and in addition it opens up a wide field of investigation in 
which other inventions and discoveries now lie hid. 2 

It is somewhat curious that throughout this controversy no 
reference was made to the fact that, in the provisional specification of 
his British patent No. 2909 of 1877, Edison uses the words ' intimacy 
of contact,' which were subsequently so generally used to describe 
microphonic action. In the first paragraph of this provisional 
specification he says : 

The vibrations of the atmosphere which result from the human 

1 Nature, xviii. 356. 

Telegraphic Journal and Electrical Review, vi. 266. 


voice or from any musical instrument or otherwise, are made to 
act in increasing or lessening the electric force upon a line by opening 
or closing the circuit or increasing or lessening the intimacy of contact 
between conducting surfaces placed in the circuit. At the receiving 
station the electric action in one or more electro-magnets causes 
a vibration in a tympan or other instrument similar to a drum and 
produces a sound, but this sound is greatly augmented by mechanical 
action. 1 

In the complete specification he states (apparently in contra- 
diction of the provisional) : 

I find that it is not practicable to open and close the line circuit 
in instruments for transmitting the human voice ; the circuit to 
the line must be always closed and the transmission be produced 
by a rise and fall of electric tension resulting from more or less 
resistance in the line. This resistance may be produced in several 
ways. I have shown several which will hereafter be named, but 
I find the most delicate to be small bunches or tufts or disks of 
semi-conducting elastic fiber, such as particles of silk and an inter- 
mediate conducting or semi-conducting material ; this device I 
call an electric tension regulator ; it is more or less compressed 
according to the vibrations of the diaphragm or tympan, and the 
electric current rises in tension as it is compressed or lessens as the 
fiber expands. 2 

In his United States specification dealing with the fibre device 
Edison says : 

I have discovered that if any fibrous material such as silk, 
asbestos, cotfon, wool, sponge or feathers be coated by rubbing 
or otherwise with a semi-conducting substance, such as plumbago, 
carbon in its conducting form, metallic oxides, and other conducting 
material, and such fiber be gathered into a tuft and placed in a 
circuit, it is very sensitive to the slightest movement. I am enabled 
not only to obtain the regulation by the greater or less pressure, 
but also to increase or decrease the extent of surface-contact between 
the particles of conducting or semi-conducting material that is 
associated with the fiber. 3 

Again, in the British specification he says : 

In some cases I make use of a variable resistance resulting from 
greater or less intimacy of surface contact such as would result 
from a disk covered with plumbago placed adjacent to a diaphragm, 
also covered with plumbago or other semi-conducting material 

1 British specification, No. 2909, 1877, p. T, lines 8-14. 

2 Ibid. p. 5, line 20. 

3 U.S. specification, No. 203,015, dated April 30 1878 (application filed 
August 28, 1877). 


so that the proximity or extent of surface contact will produce 
rise and fall of tension, the respective parts being in the telegraphic 
circuit. 1 

The variations in the extent of surface contact were the means 
of lessening or increasing the resistance in an elastic combination 
of conducting or semi-conducting material which he termed a 
' tension regulator/ though it varied the resistance rather than 
regulated the tension. 

This particular form of ' tension regulator ' was not com- 
mercially used. Superior results were obtained from ' the best 
quality of lampblack retained within a 
case.' 2 The lampblack was compressed 
into a button or lozenge, and it was this 
form of ' tension regulator ' which was 
practically used, and consequently became 
the basis of comparison. The compressed 
carbon button was in some respects a de- 
parture from the theoretical design of the 
invention because it was not sufficiently 
elastic, and whilst its efficacy was attributed 
to the variations of conductivity under 
varying degrees of compression, the re- 
searches of Hughes demonstrated that the 
results were obtained by the variations pro- 
duced upon the surfaces in contact and not 
by the mass. In this way, however, its 
operation would accord with that of the 

FIG. 34. Edison's 

Design for Transmitter digc and diaphragm above referred to in 
with Varying Surface . , ...... 

Contact. Edison s patent, which in his own words 

obtained its results ' from greater or less 

intimacy of surface contact.' The drawing illustrated in fig. 10 
of this patent and reproduced in fig. 34 is very suggestive of a 
double microphone, but the detailed description removes it from 
this category. 

This design was intended to utilise opposed batteries, the pur- 
pose being thus described : 

The battery d 2 has zinc to the line or spring c 2 , and the battery 
d 3 has copper to the line or spring c 3 . When the springs c 2 and c 3 
arc adjusted to make contact with the diaphragm equally no current 
passes to the line, but when the diaphragm is vibrated its movement 
to one side, say c 2 , causes a greater pressure upon the plumbago 
on that spring arid a lessening of the pressure on the plumbago on 
c 3 , hence the balance of the batteries c 2 and c 3 will be destroyed, 

1 British specification, No. 2909, 1877, p. 6, line 5. 2 Ibid. p. 6, line 32; 


c 2 having the advantage will send a negative current to line ; upon 
the return of the diaphragm the battery currents will again neutralise 
each other. The vibration of the diaphragm to the other side 
causes the pressure to be reversed and the battery d 3 will send a 
positive current to the line. 1 

The diagram with the above description alone might well suggest 
microphonic action, which suggestion is, however, removed when 
the description of the points of the springs c 2 and c 3 are considered. 
These points are to be 

made of compressed plumbago mixed preferably with gum rubber ; 
but any substance not liable to rapid decomposition, or the elastic 
or fibrous tension regulator aforesaid, may be used. These points 
face each other on opposite sides of the diaphragm and make contact 
with platina foil disks secured to the diaphragm. 2 

The variation in the current is clearly expected to be obtained 
through the varied resistance of the more or less elastic substance 
of the points, and not by their varied degrees of contact with the 
platinum foil discs secured to the diaphragm. 

The verdict of history needs not to be changed. Edison did 
not realise that the variation of resistance between points in contact 
would suffice for the transmission of speech, but he did realise that 
changes of surface contact in the particles composing a tension 
regulator were of value, and the attention which he gave this 
subject perhaps renders his interposition in the controversy referred 
to more reasonable than appeared at the time, especially when it is 
remembered that Hughes's claims reached him in an incomplete 

But whilst Edison with all his telegraphic and electrical experi- 
ence generally did not contemplate the possibility of a transmitter 
operating by simple variations in contact, Berliner, who obtained 
his information as to imperfect contacts at second hand, invented 
such a transmitter two or three months before the date of Edison's 
patent, and nearly a year before Hughes communicated his paper 
to the Royal Society. 

It was on April 14, 1877, that Emile Berliner deposited a caveat, 
and on June 4, 1877, an application for a patent, on a transmitter 
consisting of a diaphragm in loose contact with an electrode. Emile 
Berliner was employed in Washington at the time of the Centennial 
Exhibition, and although he had not seen the Bell exhibit he had 
heard of the transmission of speech, and had become so far interested 
in the subject as to conduct experiments. He had formed 

1 British specification, No. 2909, 1877, p. 7, lines 23-32. 

2 Ibid. p. 7, lines 17-21. 


the impression that the transmission might be more satisfactorily 
accomplished by a battery instrument than by the magneto form of 
Bell. Berliner was acquainted with the chief operator of the fire 
alarm telegraph office a gentleman named Richard and from 
him Berliner learned that in order to send an effective telegraphic 
current it was necessary to make a firm contact with the Morse 
key. Having ascertained in this way that in practical telegraphy 
there was a variation in conductivity due to a greater or less pressure 
between contacts in an electrical circuit, Berliner made two instru- 
ments, each consisting of an iron diaphragm and a steel ball. He 
connected two of them, one upstairs, the other downstairs in a 
three-storey building. He had a friend talk into the instrument 
upstairs, and he himself listened carefully downstairs and could 
plainly understand what was said. 1 He applied for a patent, and, 
the application becoming known to the Bell Company, that Com- 
pany purchased the invention and engaged the inventor in their 

The deposit of Berliner's application in 1877 gives him the 
priority in the discovery of the utility of the variable contact in 
telephone transmitters. The success of his experiment was perhaps 
in the nature of a happy accident. He did not develop or propound 
the microphone theory. But there is no disputing the fact of his 
priority of invention for a contact telephone. 

Reverting to the commercial situation as it existed between 
the Western Union and the Bell interests, it was noted that the 
introduction of the Edison transmitter gave the Western Union 
a considerable commercial advantage, and, moreover, demonstrated 
that battery instruments of some form must replace the magneto 
instrument as a transmitter. 

In the summer of 1878 Mr. Francis Blake, junior, became 
associated with the Bell Company. Mr. Blake had some scientific 
training, and was connected with the United States Coast Survey. 
Having studied the published accounts of the work of Hughes, he 
designed a transmitter on the microphonic principle, and conceived 
the idea that the electrodes should be attached to separate springs 
converging together and jointly impinging upon the diaphragm. 
The principles of the design were excellent, and of the details 
remaining the selection of suitable carbon was the most impor- 
tant. The manufacture of special carbons then in progress 
for electric lighting purposes provided the raw material, and 
a process of hardening said to have been devised by Berliner 
completed the requirements. The Blake transmitter was now 
available for public use. It was issued to subscribers, and by 

1 Telephone News, Philadelphia, Febuary r, 1911. 


its merits held control until the conditions of service had radically 

The merits of the Blake transmitter have been eloquently 
defined' by Mr. Lockwood in his ' Practical Information for Tele- 
phonists,' 1882, p. 55, and in an article on the telephone in the 
'Boston Electrical Handbook' (1904), which, though unsigned, is 
evidently by the same hand. 

Professor Hughes described the effects obtained in his experi- 
ments as being produced ' simply and solely by the direct effect of 
the sonorous vibrations,' the diaphragm having been altogether 
discarded. In the numerous applications of the microphone to the 
transmission of speech a diaphragm has always been used. The 
carbon pencil type adopted by Ader, Crossley, Cower and others 
had a diaphragm of wood placed nearly horizontal, to which the 
carbon pencils were attached ; the Blake, Hunnings, and the 
numerous successors of the latter retaining the circular diaphragm 
adopted by Edison. There is room for doubt whether Hughes gave 
a sufficiently wide meaning to the word ' diaphragm ' in its applica- 
tion to telephones. As an archdeacon was defined as one who 
undertook archidiaconal functions, so it may be reasonable to 
argue that anything which performs diaphragmatic functions is 
a diaphragm. An expert witness, Professor Silvanus Thompson, 
in the English courts maintained that ' a diaphragm is something 
which separates something from something else.' x Edison's 
original British patent contained thirty claims, but it was sustained 
only after it had been reduced by amendments to the single claim 
for the combination of a diaphragm and a tension regulator. 

The fixing of the diaphragm in its case involved considerable 
difficulty. It needed to be held firmly, yet not rigidly. Its move- 
ments should follow the movements of the air and be indepen- 
dent of any vibration through mechanical contact. Packing was 
resorted to in the form of springs or elastic substances on either 
face, but in such cases the edge remained unprotected. Edward F. 
Wilson, of Boston, conceived the idea of protecting edge and both 
faces with a single rubber band 

the internal diameter of which when not stretched is less than that 
of the diaphragm. This band is stretched over the edge of the 
diaphragm so as to form what may be termed a ' binding ' for it, 
covering the edge proper, and extending upon both faces from the 
edge towards the centre for the distance of about a quarter of an 
inch. The band should be small enough to exert in a small degree 
a pressure toward the centre of the diaphragm, so as to render the 
diaphragm slightly concavo-convex. It will be found that by the 
application of an elastic band to the diaphragm certain troublesome 

1 Electrical Review, London, x. 638. 


overtones are avoided which are present when the edge proper is 
left bare. 1 

Wilson's patent, from which the above is quoted, is an example 
of many very simple applications of great practical value. The 
rubber ring around the diaphragm first applied to the Blake trans- 
mitter is a feature which has been retained. 

In the United States the BeU patent controlled the electrical 
transmission of speech however obtained, so that Edison's was quite 
subsidiary. The Blake and other carbon transmitters employing 
carbon as a contact material were, however, generally regarded as 
being subordinate to the Edison invention of the carbon transmitter 
and of the contact transmitter of Berliner. 

When it is considered that the essentials of a transmitter were 
accurately recorded by Edison as two the diaphragm and the 
' tension regulator ' and when it is considered also that the dia- 
phragm does not lend itself to many changes, it is matter for wonder 
that so many variations of the ' tension regulator ' could be made 
as to have permitted almost every new entrant in the telephone 
field an opportunity for a new christening. Some, of course, had 
detail merit, but most were modifications covering no new principle 
and effecting no improvement in results. In consequence they need 
no mention, and at this stage one other only will be referred to. 

The Rev. Henry t Hunnings, of Bolt on Percy in the county of 
York, a clergyman of the Church of England who combined with 
theology a taste for natural philosophy, devoted some attention 
to the telephone, and as a result of his experiments, which, he said, 
' were made in total ignorance of any experiments having been 
made by any other worker,' z he invented a transmitter which is 
thus described : 

A front vibrating diaphragm, composed wholly or in part of 
suitable metal, such as, preferably, platinum, silver, ferrotype- 
iron, tinned iron, el cetera. In close proximity to the aforesaid 
vibrating diaphragm is fixed a disc of brass or other suitable 
metal, the intervening space being filled with carbon in the form 
of powder to the depth of about one-sixteenth of an inch. The 
aforesaid vibrating diaphragm and the fixed disc of brass are con- 
nected, respectively with the opposite poles of a voltaic battery. 
The whole may be secured in a box of suitable non-conducting 
material, with a mouth-piece, if desired. 3 

He remarks that 

the details can be indefinitely varied, the great feature being in the 

1 U.S. specification, No. 250,616, December 6. 1881 (application filed 
May 21, 1879). 

2 Electrical Review, London, x. 350. 

* British specification, No. 3647, September 16, 1878. (U.S. equivalents, 
No. 246,512, August 30, 1881, and No. 250,250, November 29, 1881.) 



use of carbon used in a state of fine loose powder, not in any way 
compressed or consolidated, as I find the loose particles of con- 
ducting matter to be most delicately sensitive to sonorous vibrations. 1 

Fig. 35 is a reproduction of the drawing forming part of the 
patent specification. 

Hunnings, or at any rate his patent agent, was familiar with 
the Edison transmitter and the Hughes microphone, but, so far as 
the patent specification gives any indication, does not appear to 
have been acquainted with the whole of Hughes's paper, or of his 
reference to the utility of carbon or metals in a finely divided state. 

The Edison, Blake, and Hunnings 
transmitters, together with Hughes's 
discovery of microphonic effects, all 
influenced subsequent developments 
referred to in later chapters. The 
Hunnings instrument played but little 
part at the period to which we are 
now referring. An attempt was made 
by a rival Company in London to 
establish its independence of the Edison 
patent an attempt which succeeded 
on a technical point in the court of 
first instance only to fail on appeal. 
The utility of the Hunnings loose carbon 
was to be developed later. But the 
Blake played a part of immediate 
importance. It gave the Bell Company 
an instrument which, though not as 

loud as the Edison, was its superior in FlG 35 Running's Trans- 
clearness of articulation, in reliability, mitter, as illustrated in 
and in durability. The following para- patent, 
graphs are taken from a circular entitled ' Directions for setting 
up the Edison Transmitter Boxes ' : 

Battery (either carbon or gravity) is needed at each station, and 
is used alternately as local and main. It is used as local when 
talking, and as main to call with. 

The battery at all the stations is left on the main circuit for 
calling, except when conversation is being carried on. The switch 
changes the battery from the local circuit to the main line. The 
battery of one station remaining in circuit does not interfere with 
talking between any other two stations. 

For short lines one cell of carbon or two cells of gravity battery 

1 British specification, No. 3647, September 16, 1878. (U.S. equivalents, 
No. 246,512, August 30, 1881, and No. .250,250, November 29, 1881.) 


are needed at each station. These are enough for speaking purposes 
for any length of line, but for long lines more cells are needed for 
calling. These may be placed at any point in the main circuit. 

The carbon button of the Edison transmitter is delicate and 
liable to disarrangement from a slight jar. The instrument should 
therefore be transported with great care, and the box be fastened 
to the wall before the transmitter is put in place. 

The adjustment of the carbon button was also a work of difficulty 
and delicacy. The Blake suffered much less in these respects. 
The electrodes were mounted on the short arms of a lever. A 
screw played on the longer arm permitting the most delicate adjust- 
ment of the electrodes in their relation to the diaphragm. The 
carbon was so hard as to be free from danger, and there was no need 
either to pack or transport the transmitters as if they were eggs. 

The production of such an instrument was of the utmost import- 
ance to the industry, but to the Bell Company its timely arrival 
meant the saving of the situation in the commercial competition 
with its rival. 



Ax that period of Philipp Reis's career which is of especial interest 
to us, he occupied a post in Garnier's Institute at Frankfort-on- 
Main. A teacher by profession, he was in the primary sense of the 
word an amateur of science. He was interested in new discoveries 
and developments, and acoustics had a special attraction for him. 

Chronologically his work should have been referred to at an 
earlier stage, but since it had no direct influence upon the 
development of Bell's invention, its consideration has been reserved 
for this chapter, because such consideration is facilitated by the 
preceding chapters. 

So far as can be judged by his Telephone Memoir, he seems to 
have presented his subjects in a practical manner, avoiding exaggera- 
tion, yet not oblivious to a proper recognition of any contributions 
he might make to scientific progress. Brevity, though generally 
commendable, was in his case perhaps something of a misfortune, 
for he desired to be understood, and if he could have foreseen the 
diverse interpretations which were to be placed in later years upon 
his aims, his explanations, and his results, he would undoubtedly 
have elaborated his illustrations and chosen his words with such 
care that it would not have been possible to misunderstand. 

Reis invented his apparatus which he called The Telephone, 
he exhibited it in operation and explained the theory before a 
scientific society to which he belonged, and the apparatus was 
described together with the degree of success attained with it 
in scientific and other publications. In 1863 the Reis telephone 
was exhibited before the British Association by Mr. Ladd. He, 
obtained it from the authorised manufacturer, and received an 
autograph letter from Reis describing its construction and the 
method of its operation. Its exhibition was recorded in the ' Trans- 
actions ' of the Association without any reference being made to 
speech. It is curious to note also that the name of the inventor 
was not given. Dr. Ferguson in his book ' Electricity,' published 



in 1866 (p. 257) describes the mechanism with precision, explains 
the operation of the transmitter as being that of a circuit breaker, 
the sound in the receiver as being due to the ' magnetic tick,' 
and the combined result as the transmission of pitch only. Regard- 
ing the reproduction he says : ' The note is weak, and in quality 
resembles the sound of a toy trumpet.' 

It is natural that any developments in science or invention 
should revive interest in previous work on similar lines. When 
therefore Bell succeeded in transmitting speech, interest in Reis's 
work, which had been dormant, was revived ; but it was not until 
Bell's telephone had been in existence long enough to demonstrate 
its practical and financial value that it was alleged Reis had 
previously transmitted speech. 

Lack of novelty is fatal to a patent, and the effort to prove 
prior invention is therefore an almost invariable feature of the 
defence in a patent action. There were several such actions about 
the telephone, and the claims put forward for Reis as a prior 
inventor figured in most of them. 

These claims were investigated by Federal Courts in Boston, 
New York, New Jersey, Philadelphia, at Pittsburg, New Orleans, 
and by the Supreme Court of the United States at Washington, 
who one and all decided that Reis was not the inventor of a telephone 
in the sense of inventing an instrument which would transmit 
articulate speech. English courts, in the case of the United Tele- 
phone Co. v. Harrison, Cox-Walker & Co., 1 also decided that the Reis 
instruments and publications did not anticipate Bell. 

But the work of Reis, nevertheless, was meritorious, and in giving 
a brief account of it the opportunity may be taken to explain the 
misunderstanding of that work which underlay the claims per- 
sistently submitted to legal judgment and as persistently rejected. 

Like the more modern telephones which have been described, 
Reis's apparatus consisted of two parts a transmitter and a 
receiver. The receiver had to reproduce the sounds, and whatever 
results were expected must necessarily be limited to the effects 
which the apparatus was known to be capable of producing, or that 
the inventor contemplated it was capable of producing. 

The receiving apparatus adopted by Reis was that used to 
illustrate the magnetic sounds known since 1837 as the ' Page 
effect.' Dr. Ferguson in his ' Electricity ' gives an explanation of 
this effect as follows : 

Magnetic Tick. When an iron rod is made to rest on a sounding 
board, such as the body of a fiddle, and placed in the centre of a 
powerful coil, each time the current is broken a distinct tick is 

1 Telegraphic Journal, x. 381. 


heard from the rod. If a file be placed in the circuit, so that a 
wire when it slides along will alternately close and open the circuit, 
the rasping noise of the wire sliding along the file will be distinctly 
rendered by the rod, each interruption giving rise to a tick ; the 
series of ticks being in the same order exactly as the series of noises 
at the file. 1 

In his own record of the experiment as originally carried out, 
Page in Silliman's Journal, July 1837, said that ' when the contact 
is made the sound emitted is very feeble, ; when broken, it may be 
heard at two or three feet distance.' 

It will be seen that Ferguson refers only to the broken current 
as producing the tick. Reis adopts the theory that ' at every closing 
of the circuit the atoms of the iron needle are pushed asunder from 
one another,' and ' at the interruption of the current the atoms 
again attempt to regain their position of equilibrium.' The ' inter- 
ruption ' here is clearly the breaking of the circuit. 

Reis makes no claim to have discovered in the receiver any new 
qualities, nor beyond the hypothesis referred to on p. 139 does 
he advance any new theory in the method of its operation. It had 
produced sounds before by repeated ticks. Reis gives us no ground 
for assuming otherwise than that he not only expected it to operate 
by repeated ticks, but also that the ticks would be produced by 
breaking the circuit. 

While the receiver was adopted as a ready-made appliance, 
the transmitter was designed by Reis himself. Both are illustrated 
in fig. 36. 

Before Reis's time the knitting needle contrivance illustrating 
the Page effect had only been operated by a mechanical circuit 
breaker of some sort, of which the wire and file mentioned by 
Ferguson is one example. The transmitter of Reis permitted the 
operation of the circuit breaker directly by air vibrations. 

Its construction was ingenious. The metal patch in the centre 
of the diaphragm was one electrode. The other electrode (in the 
standard model) was one foot of a tripod resting normally on the 
centre of the diaphragm in contact with the lower electrode. The 
other two feet of the tripod rested on the box in suitable ' cups ' 
or recesses. In one of the cups there was placed ' a little drop of 
quicksilver,' evidently for the purpose of ensuring conductivity 
under movement. The movement, its object, and its result are 
very clearly described. When the diaphragm is set in vibration, 

at the first condensation the movable electrode will be pushed back. 
At the succeeding rarefaction it cannot follow the return vibration 
of the membrane, and the current . . . remains interrupted . . . 

1 Electricity, Ferguson, 1866, p. 182. 


until the membrane, driven by a new condensation, presses the 
lower electrode against the upper electrode once more. In this way 
each sound wave effects an opening and closing of the current. 1 

Early publications on the speaking telephone include descriptions 
of Reis's apparatus without, as a rule, an analysis of the operation 
of the various parts, and the descriptions are sometimes incorrect. 
Prescott (1878) (p. 147) adopts a translation of the article in 
Bottger's ' Polytechnical Notezblatt,' 1863, in which both electrodes 
are described as ' fastened ' to the wood, though the ' hopping ' 

FIG. 36. Reis Transmitter and Receiver (from 'Electricity,' Ferguson). 

character of the upper electrode is referred to. Du Moncel (1879) 
(p. 12), after mentioning the diaphragm with its platinum disc, 
says that above this ' a metallic point c was fixed, and this together 
with the disc constituted the contact breaker.' 'The Modern 
Applications of Electricity ' by Hospitalier, translated by Julius 
Maier (1882), says (p. 297) : 

Each time that the membrane is raised the point will touch 
the disc, and a current will be established. It is, on the contrary, 
interrupted when the membrane comes back into rest. 

The description leads to the inference that the upper electrode 
is fixed, though the explanation of the illustration on the next page 

1 Reis, Frankfort lecture. 


mentions it as a ' movable lever touching the membrane.' Dr. 
Maier collaborated with Sir William Preece in the authorship of 
' The Telephone ' (1889), in which the foregoing description from 
' Modern Applications ' is substantially reproduced without the 
explanation to the illustration. 

Attention has been called to these references because it is 
not possible to follow Reis's ideas and appreciate his work without 
recognising that his upper electrode was not fixed. Dr. Ferguson's 
description is more detailed and more accurate a tripod rested 
with two feet on the frame and one foot on the diaphragm. The 
circuit was normally closed by the tripod resting on the disc. When 
the diaphragm became agitated the circuit was interrupted in 
accordance with the rate of movement of the' diaphragm. Thus 
the transmitter conformed to the requirements of the receiver in 
producing ' interruptions ' or breaks. 

It was on October 26, 1861, that Philipp Reis delivered before the 
Physical Society of Frankfort-on-Main his lecture ' On Telephony by 
the Galvanic Current ' : on November 16 he gave ' An Explanation 
of a New Theory concerning the Perception "of Chords^ and'' of 
Timbre as a Continuation and Supplement of the Memoir on the 
Telephone.' These lectures were combined and printed in the 
records of the Society * under date of December 1861. Though 
recorded as one paper it was, as we have seen, delivered in two 

In our quest for Reis's meaning it would have been a great help 
to know the exact portions of his paper which were delivered 
on the respective dates. The first was certainly a lecture on 
the telephone, and may be presumed to have included so much 
of the description and the theory as was necessary to understand 
it. The second was an exposition of a new theory on the perception 
of chords and of timbre, to which a table of curves presented with 
the paper seems to be a natural illustration. 

There are other curves intended specifically to illustrate vowel 
tones, and these, it may be presumed, were a part of the first (or 
telephone) paper. What Reis meant by the vowel curves is 
fully explained by analogy, so that there is no need for conjecture. 
The more extensive illustrations of compound curves which, it 
may also be presumed, were a part of the second (or perception of 
chords) paper, are not so clearly described. 

It is these last-mentioned curves which underlay the claims 
made for Reis as the inventor of the telephone. They have been 
applied to speech with a meaning that was not prevalent in Reis's 

1 The record of the lecture is given in full in Philipp Reis, Inventor of the 
Telephone, by Silvanus P. Thompson (London : E. & F. N. Spon, 1883), together 
with other contemporary documents. 



day. Reis made no reference to these curves in connection with 
speech. They were used as musical illustrations, analysed to the 
extent of four notes, and it is possible that they were adopted by 
Reis without any such perception as has been ascribed to him. 

It may be recalled that the theory of compounded tones, an 
elaborated exposition of sympathetic reinforcement, and the 
analysis of compounded tones by resonators were enunciated in 
Helmholtz's lecture at Bonn in the winter of 1857, ' On the 
Physiological Causes of Harmony in Music.' 1 

Where two writers are dealing with the same subject it is inevit- 
able that there must be some points of similarity, and a common 
origin is not necessarily to be inferred from such similarity. But 
Helmholtz was known to be an original investigator in the field 
of physics, a great scientist and a popular lecturer, whilst Reis was 
an interested student as well as a capable teacher, one who (in his 
own words) had, some nine years previously, ' a great penchant 
for what was new but with only too imperfect knowledge of 
physics.' 2 

In the preface to the English translation of Helmholtz's 
' Popular Scientific Lectures ' (1873) it is said that the Bonn lecture 
had not previously appeared, so that it is not possible for Reis 
to have read this lecture as a whole. Whether he heard it or 
whether any partial publication of it may have been made, can 
only be matter of conjecture. 

Reis was evidently not a man who would put forward as his 
own the idea of another, but it is to be noted that the explanation 
of the new theory whiph the curves served to illustrate is not stated 
to be his own theory, though some support might be found for such 
a suggestion in his remark that the correctness of ' my views with 
respect to the curves representing combinations of tones may 
perhaps be determined by the aid of the new phonautograph.' It 
is only from internal evidence, and perhaps not unreasonable 
i nf erence, that it can be suggested that Reis was probably familiar 
with some portions of Helmholtz's Bonn lecture. 

Helmholtz remarked that the theory which he enunciated in this 
lecture in 1857 ' will perhaps seem new and singular.' The title of 
Reis's second lecture was ' An Explanation of a New Theory con- 
cerning the Perception of Chords and of Timbre as a Continuation 
and Supplement of the Memoir on the Telephone.' 

Helmholtz told his audience they must ' conceive the air of 
a concert hall or ball-room traversed in every direction, . . . by a 
variegated crowd of intersecting wave-systems,' and, after describing 
his sound curves and other illustrations, asked, ' Now, what does the 

1 Popular Scientific Lectures, Helmholtz, p. 61. 
* Reis, Frankfort lecture. 


ear do ? Does it analyse this compound wave ? Or does it grasp 
it as a whole ? ' Developing his theory of Corti's arches he sug- 
gested the probability that it was analysed into its constituent 
parts by the individual fibres. He further drew a distinction 
between the audible sensation and the intellectual conception. 
' We have, as it were, to distinguish between the material ear of 
the body and the spiritual ear of the mind.' Reis said : 

How does our ear take cognizance of the total vibrations of all 
the simultaneously operant organs of speech ? Or, to put it more 
generally : How do we perceive the vibrations of several bodies 
emitting sounds simultaneously ? . . . The function of the organs 
of hearing, therefore, is to impart faithfully to the auditory nerve 
every condensation and rarefaction occurring in the surrounding 
medium. The function of the auditory nerve is to bring to our 
consciousness the vibrations of matter resulting at the given time, 
both according to their number and their magnitude. 1 

In this Bonn lecture Helmholtz illustrated two simple sounds by 
their appropriate curves and their combination in a compound 
curve. Reis adopted a similar method in his Frankfort lecture, 
but carried the illustrations further. There is some importance 
in the consideration of the question whether Reis from independent 
thought created his curves or whether he adopted them from 
Helmholtz. It is without doubt that Helmholtz publicly used the 
illustration four years before Reis's lecture. 

Our interest lies in realising, if possible, Reis's meaning of the 
curve, and in this endeavour to discover what was in Reis's mind we 
must consider the whole of his paper, the theory which he propounded, 
and the apparatus by which he proposed to convert it into practice. 

In this consideration either of two methods may be followed. 
One may start with the description and illustrations which are 
capable of varied interpretation, according to the extent of 
knowledge assumed, and then proceed to consider the apparatus 
with a view to seeing how far it may meet those interpretations. 
Or one may argue backwards from the apparatus. The latter 
affords the more ready and probably the more accurate method 
of realising Reis's conception. 

When he comes to define the transmission Reis says : 

As soon, therefore, as it shall be possible at any place and in 
any prescribed manner to set up vibrations whose curves are like 
those of any given tone or combination of tones, we shall receive 
the same impression as that tone or combination of tones produced 
upon us. 1 

1 Reis, Frankfort Lecture. 

K 2 


He anticipates that the results at which he aims will be secured 
if he transmits number and strength. This is clearly expressed in 
his explanation of the reproduction of the vowels. He alludes to 
' The researches of Willis, Helmholtz, and others/ showing that 
vowel sounds can be artificially produced by causing the vibrations 
of one body to reinforce those of another periodically.' He explains 
this operation by the use of an elastic spring set in vibration by the 
thrust of a tooth of a cog-wheel : 

The first swing is the greatest, and each of the others is less 
than the preceding one. After several vibrations of this sort 
(without the spring coming to rest) let another thrust be given 
to the tooth ; the next swing will again be a maximum one, and 
so on. The height or depth of the sound produced in this fashion 




FIG. 37. Vowel Curves from Reis Lecture. 

depends upon the number of vibrations made in a given time ; 
but the quality of the note depends upon the number of varia- 
tions of amplitude occurring in the same time. 1 

He uses curves in illustration. These curves are expressly 
applied to speech, and it is from these that we must form our 
impression of what Reis conceived to be necessary. He says : 

Two vowels of equal pitch may be distinguished from each 
other somewhat after the manner represented by the curves (i), (2) ; 
while the same t9ne devoid of any vowel quality is represented 
by the curve (3). 

Our organs of speech create the vowels probably in the same 
manner by a combined action of the upper and lower vocal chords, 
or of the latter and of the cavity of the mouth. 1 

The two different vowels illustrated by Reis are assumed to be 
obtained by vibrations of the same period with a reinforcement in 

1 Reis, Frankfort lecture. 


the first case after every fourth beat, and in the second case after 
every sixth beat, there being a gradual loss of power in either case 
between each reinforcement. 

This being Reis's conception of the differences in vowel tones, 
he expected his apparatus to perform the function which he had 
outlined for the ear : 

The greater the condensation of the sound conducting medium 
at any given moment, the greater will be the amplitude of vibration 
of the membrane and of the ' hammer/ and the more powerful 
therefore the blow on the ' anvil ' and the concussion of the nerves 
through the intermediary action of the fluid. 1 

Since the parts referred to as constituting the mechanism of 
the ear were reproduced in his transmitter he expected to obtain 
corresponding results. It was to meet this requirement that he 
devised the peculiar form of circuit breaker in his transmitter. 

The ingenuity of the arrangement may be better appreciated by 
considering the alternative. The first idea would naturally be to 
follow the vague suggestion of Bourseul and provide a vibrating 
diaphragm which should make and break contact with a fixed 
electrode. Such a plan could not effectually allow for the variations 
in the amplitude of the vibrations which Reis had clearly in mind. 
He reversed the order, and instead of arranging for the making 
or breaking of contact by the extended diaphragm operating as 
a sensitive key upon a separate stud he provided a means of breaking 
contact when the agitation of the diaphragm threw upwards the 
movable electrode normally resting on it. The upper electrode 
was a point and the lower electrode a disc. In a drawing the 
former would be regarded as a hammer and the latter as 
an anvil. The reversal of the ordinary operation which Reis 
carried out may be comprehended readily by assuming that instead 
of a hammer striking an anvil, the anvil by a rapid movement threw 
up the hammer. This would occur with a condensation, and in the 
subsequent rarefaction the diaphragm would retreat more quickly 
than gravity could operate upon the upper electrode, thus effecting 
the alternate separation and contact. When, in any compounded 
sounds, the vibrations coincided, the condensation would be of 
greater strength, the ' blow ' would be more powerful. This is 
explained when he says : 

Moreover, the strength of this tone [i.e. this tick] is propor- 
tional to the original tone [i.e. original vibration], for the stronger 
this is, the greater will be the movement of the drum-skin, the 
greater therefore the movement of the little hammer, the greater 
finally the length of time during which the circuit remains open, l 

1 Reis, Frankfort lecture. 


We may interrupt the quotation here to note Reis's thoroughness 
in his desire to follow the process through. Since it is required to 
reproduce the sounds, he has to explain how these operations in the 
transmitter are reflected in the receiver. He continues : 

and consequently the greater, up to a certain limit, the movement 
of the atoms in the reproducing wire (the knitting needle), which 
we perceive as a stronger vibration, just as we should have perceived 
the original wave. 1 

At the beginning and at the end of the lecture the same idea 
is expressed. The function of the auditory nerve was denned in 
his introduction as the bringing of the vibrations to the conscious- 
ness ' both according to their number and their magnitude,' and 
in his conclusion he says that his apparatus permits the transmission 
of ' the number and the strength ' of the vibrations. Consequently 
he expected to produce varied qualities of sound by trans- 
mitting detached currents which should vary not only with 
the number of vibrations in a given time (producing pitch), but 
which should also vary with the force of individual vibrations, 
by which he expected to reproduce timbre because that was his 
conception of the audible perception of timbre. Whether he 
obtained on the line and in his receiver the expected varied 
effects from varied amplitudes is open to question. But it suffices 
to indicate that such effects were involved in his theory and that 
the production of such effects was considered to be provided for 
in his apparatus. In the ear he assumed the operation of number 
and magnitude. In his apparatus he thought he had provided for 
number and strength. The imperfections in operation which he 
recognised he attributed to defects in the transmission or repro- 
duction. What he expected to attain and conceived to be 
sufficient may be graphically represented by circles of differing 
size. Fig. 38 indicates in this form the two vowels and one simple 
sound illustrated by Reis's curves in fig. 37. The sizes of the 
circles represent the varying strength of the blows assumed to be 
effected by each vibration. The detached currents transmitted to 
line were in the reverse order the strongest blow was expected 
to convey the shortest current. In the reproduction of the 
vibrations in the receiver the variations in the degrees of loudness 
were assumed to be effected by the longer or shorter period afforded 
for the readjustment of the molecules in the knitting needle. 

If it be assumed that Reis's complex curve was intended to 
be a graphic representation of several separate vibrations it will 
be in accord with the knowledge of the period, and we may read 

1 Reis, Frankfort lecture. 


his paper without doing violence to any of his words. If it be 
assumed that he conceived the necessity of transmitting the whole 
of his complex curve with all its sinuosities in the manner which 
after 1876 was expressed by the term ' undulatory current/ we 
must not only do violence to his language but we must assume also 
that he had made important discoveries in the operation of 
both transmitter and receiver without making any reference 

It is no compliment to Reis's memory to assume that he omitted 
to record any new discovery which he may have made, or that 
there was any looseness of expression in his use of common terms. 
When lecturing on a new theory regarding the perception of chords 
it is hardly to be expected that he would refrain from describing 
a new discovery of his own if he had made one. The discovery 

(3) & 

FIG. 38. 

of the microphonic effect must not be confounded with a general 
knowledge that current was subject to some variation according 
to the degree of pressure on a Morse key or similar contact maker. 
Such an effect was on record though not generally known. The 
discovery that such an effect would serve to modify a current with 
the almost infinite variety required for the transmission of speech 
vibrations is a very different matter. 

Throughout his papers there is not only no indication that 
Reis knew anything about the effect of a variation in resistance 
between electrodes in imperfect contact, but there are very 
definite statements that the contacts of the transmitter must 
make and break. Moreover, his receiver was a contrivance 
which had been used, and only used, to produce sound resulting 
from the molecular disturbances in the core of an electro- 
magnet as the result of repeated breaks in an electrical circuit. 
If Reis had discovered any new method of tone production 
from it, we may infer that he would have been glad to say so. He 
does not say so, but he gives an exposition of the magnetic theory 
of its operation, the basis of which he quotes from Muller-Pouillet, 


' Lehrbuch der Physik.' 1 There is in his description of such theory 
no indication that he had discovered and sought to use a new 
property not previously known or used. Reis clearly expected 
to obtain from his needle a series of ticks such as had been 
previously obtained, and did not regard it as possible to obtain 
sound from it in any other form. 

Scientists learned with expressed surprise many years later, 
and after the electrical transmission of speech had been effected 
by other means, that variations in contact pressure would serve 
to transmit articulate speech. At an earlier stage they were still 
more surprised at the demonstration that the variations in a magnet, 
operating upon a diaphragm armature, would reproduce articulate 

The contemporary records of Reis's demonstrations have been 
alluded to. In the course of the litigation other demonstrations 
were made. The results were summarised by Mr. Dickerson as 
follows : 

When the Dolbear case was tried, Mr. Dolbear, who was one 
of the early admirers of Bell, and who had published a book which 
lauded him (but who joined the infringers afterwards), said that he 
could make this Reis talk. We told him to go ahead and make it 
talk. We only asked the privilege of being present, and having 
a stenographer at each end, so that one man could take down what 
was said and the other could take down what was heard. Mr. 
Dolbear and Mr. Buck were the operators. They had several 
instruments the box kind and cone kind. Their box set they 
had gone to Germany for and thought it the best they had. They 
had been experimenting with it for months. They met us and 
tried to make it talk. They talked over a thousand words into 
it on two different days. They used familiar phrases, such as ' Mary 
had a little lamb,' ' Hello,' ' How do you do ? ' ' How is this for 
high ? ' and all that sort of thing. This apparatus will give you 
the general rhythm of the sound, and from it a man can guess once 
in a while what he hears. Out of more than a thousand words they 
guessed at fifteen or twenty, and they guessed more than half wrong 
and less than half right. . . . Edison said, in one of his depositions 
about a telephone which he had thrown away as worthless, ' When 
you knew what it was the man was saying it sounded awful like.' 
... If you get a rhythm, identifying it with a phrase you are 
accustomed to, you may guess right. Suppose you are at the 
instrument, and the phrase ' Mary had a little lamb ' was sent to 
you, and your assistant had used it many times, and it had been 
sent many times, you might be able to think you recognised it by 
the length of the words and the rhythm. And so in that way they 
guessed at fifteen words out of the thousand ; but of those they 

1 Vol. ii. p. 304, 5th edition. 


guessed, eight were wrong and seven right. That is the best that 
has ever been done. 1 

These are the words of an advocate. The facts which he refers 
to were in evidence before the Court. The conclusions he forms 
from those facts are his own. The material interests which he was 
defending were great. Those interests have long since lapsed ; 
the scientific interest alone remains. But independent effort to 
discover Reis's mental conception, and careful consideration of 
the contemporary evidence available as to the operation of his 
instruments, fail to modify the contention of the advocate. Whether 
the Reis instruments ever conveyed from talker to listener a sound 
which could accurately be called speech can never with certainty 
be known. The probabilities against it are so great that over- 
whelming evidence is necessary to support the contention that 
speech was transmitted. And the contemporary evidence does 
not indicate more than that sounds suggestive of words were sent. 
All the recorders were friendly recorders and were disposed to make 
the best of the apparatus. Legat had formed the highest opinion 
of its importance, was the most enthusiastic, and was probably 
the most competent demonstrator after Reis himself. Legat's 
report shows that chords and melodies ' were transmitted with 
marvellous fidelity/ single words ' were perceptible more indistinctly ' 
but here also modulations and exclamations ' attained distinct 
expression.' This accords with what might be expected from an 
examination of the theory and the instruments. In cases where 
the ground tone formed the prominent characteristic of a word 
the word was suggested, but in any case where overtones became 
an essential to identification the characteristic feature was lacking 
and the identity was lost. Such a conclusion is not to be modified 
by the assumption that the Reis transmitter occasionally operated 
as a microphone. The possibility even the probability that 
it did so act momentarily and accidentally must be recognised, 
but if the transmitter did by accident send an undulatory current 
out upon the line it could not be converted into an audible sound 
with the receiving mechanism provided. At the receiving end 
also we must not be misled by the possibilities of a knitting needle 
and a coil of wire as known since March, 1876. We must remember 
that Reis was not looking for a still small voice. The theory 
upon which he founded his experiments and the circumstances 
under which he carried out his demonstrations indicate that his 
only hope of sound from the knitting needle was the tick arising 
from the broken circuit. Sounds produced by aggregate ticks 

1 Mr. Dickenson's argument, American Bell Telephone Co. v. National 
Improved Telephone Co.. Circuit Court, U.S., March u, 1886. p. 113. 


were loud, so loud that at the first public demonstration he was in 
a position to make them audible ' .to the members of a numerous 
assembly/ The volume of sound necessary for this purpose shows 
that it was obtained by ticks, and the expectations of such volume 
would prevent the discovery of the extremely feeble sound (if any) 
resulting from undiilatory currents. 

The scientific world adopted the view of Reis's work, which 
is expressed by Dr. Ferguson. 1 The practical results had not 
justified any other view, and it was perhaps too readily assumed by 
some that his aims had been no higher than his results. In what 
may be called the controversial period much w x as made of the 
phrase ' reproduction of tones/ it being assumed that Reis had 
no higher aim than melodic reproduction. It was also sometimes 
claimed that because Reis provided his instruments with a Morse 
key and sounder he never intended to transmit speech. That 
Reis started out with the intention of transmitting speech is more 
than doubtful, but that he essayed the transmission of speech is 
clear from his own words. That he failed is equally clear from his 
own records. 

Reis's work had no direct effect on the invention of the speaking 
telephone, for Bell fortunately went on entirely independent lines 
and without any reference to the prior work of Reis. It is on record 2 
that Legat's report of Reis's apparatus was submitted to Edison in 
or about the month of July 1875, and formed the starting point of 
his telephonic experiments. It is perhaps on that account that 
Edison in his 1877 specification relates as a new discovery 
' that it is not practicable to open and close the line circuit in 
instruments for transmitting the human voice/ 3 Bell started with 
the realisation of that necessary fact. 

The difference in the results of Bell and Reis is plain enough, 
but the difference in the methods is plainer still. Between an origina- 
ting sound and an attempted reproduced sound Reis placed an 
electrical apparatus to obtain a broken circuit, and Bell placed an 
electrical apparatus incapable of producing a broken circuit, but 
adapted to the production of an electrical current of an undulating 
character, following precisely the curve of the originating sound. 
The methods differed because the conceptions differed. 

It is not because the world was unprepared for a ' phonic tele- 
graph/ nor because Reis was a comparatively obscure experi- 
mentalist, that his apparatus came down to a later generation 
as a musical telephone. It is because the vital principle of a 
talking instrument was lacking. 

1 P. 126. 

1 Bell's Electric Spfaking Telephone, Prescott, 1884, p. no. 
8 Chapter XI. p. 117. 


In reputation Reis has suffered somewhat from the claims made 
on his behalf. Those who saw in him the inventor of the speaking 
telephone claimed his transmitter as a microphone. Those who 
opposed such claim were content to demonstrate that he produced, 
and only intended to produce, make-and-break apparatus. 
Between these two extremes he has failed as yet to receive the 
credit to which he is really entitled. The individual breaks ' hand- 
claps/ Clerk Maxwell described them have been assumed to be 
of equal value. But Reis contemplated a differing value in the 
individual units hand-claps still, but of differing degrees of force. 
As already remarked, the capacity of his apparatus to carry his 
theory into practice is open to question, and it may be that the 
historians have correctly described what the apparatus did, but 
they failed to record what Reis hoped that it would do. His theory 
on the acoustic side, though incorrect, was not unreasonable with 
the information then available. On the electrical side his explana- 
tion of the operation of the receiver needle adopted from Muller- 
Pouillet was only an electro-magnetic reason for its performing the 
same simple function as the spring in Savart's wheel, and thus by 
responding to variations in number producing pitch. An extension 
of this explanation to account for the variations in strength was 
based on the hypothesis that a longer period of rest in the needle 
gave greater power to the subsequent tick. Some explanation of 
this sort was necessary to make the assumed electrical operation 
conform to the acoustic theory. Its accuracy is doubtful, but 
whether accurate or otherwise is of no practical importance in 
the consideration of speech transmission. The simple purpose of 
Reis's explanations being overlooked, more abstruse reasons 
suggested themselves and contributed materially to the misunder- 
standing upon which the Reis claims were based. The application 
of the same acoustic theory is observable in modifications of the 
apparatus which were exhibited by Legat. That Reis received 
little encouragement so far as speech is concerned may be taken 
for granted, but there was no lack of recognition for that which he 
actually accomplished. Fourteen years later the real inventor of 
the telephone, as we have seen, had to hide his hopes and bide his 
time because he realised that neither the scientific nor the financial 
world would regard the expectation of the electrical transmission 
of speech as other than a dream. That Reis failed does not detract 
from the credit due to him for his efforts in a direction which no 
successor had the temerity to pursue until Bell, with a courage 
prompted by wider knowledge, independently attacked the problem, 
achieving results which were immediately recognised, and which 
eventually revolutionised the methods of communication through- 
out the world. 



WHILE the consideration of Reis's work was felt to be most con- 
veniently undertaken in the preceding chapter, we have to recognise 
a certain disadvantage in the interruption thereby effected in 
the record of practical work, and in the recital of the means adopted 
for carrying on an effective service. Returning, then, to the 
practical stage we may recall that before the production of battery 
transmitters or microphones, the telephone had been applied to 
a new service which called for the development of accessory appar- 
atus of new types. With the introduction of these battery trans- 
mitters the new service became capable of great expansion. But 
before considering the developments in this direction some attention 
must be given to the more important of the accessories call bells 
and switchboards which contributed to make the new service 
possible. Only the earlier forms of call bells will be noted in this 
chapter and of switchboards in the next. 

Transmitters and receivers were now in existence by whose 
means speech could be transmitted and received with clearness 
over long lines, but the ability to talk carried with it the need to 
send a calling signal over equal distances. 

Bell's original circular had intimated that the telephone itself 
would serve for the call, since ' any person within ordinary hearing 
distance can hear the voice calling through the telephone '; but it 
also anticipated occasions when that still small voice might be 
insufficient, for it was added : ' If a louder call is required, one can 
be furnished for $5.' The apparatus by which the louder call was 
to be effected was that for which Thos. A. Watson submitted on 
December 5, 1877, an application for a patent, in which he says : 

In using a system of electric telephones it is necessary to provide 
some means for producing a sound at the distant telephone station 
loud enough to attract the attention of peisons at a distance from 
the telephone. 



My present invention provides one means for doing this by 
causing an intermittent current of electricity of high intensity to 
pass through the line wire and the distant telephone. For producing 
such current I make use of an ordinary induction coil combined 
with a galvanic battery and a rheotome, for rapidly interrupting 
the current. 1 

In this case the telephone was still used as the source of calling 
sound at the receiving station. 

F. A. Gower combined a ' musical instrument ' with a telephone, 2 
which was introduced into Europe, and is thus described by Du 
Moncel : 

The instrument can itself give a very loud call by only breathing 
into it instead of speaking. 

For this purpose a small oblong opening is made in the 
diaphragm at a half diameter from its centre, and behind this the 
reed of an harmonium is applied to a square copper plate fixed on 
the diaphragm itself. On using the bellows the expelled air passes 
through this little hole and, on reaching the reed, sets it into 
vibration, and produces a sound of which the acuteness depends 
on the condition of the vibrating plate. 3 

Gower's telephone had a flexible tube to speak into, and this 
was considered to be an advantage when using the vibrator call. 
In describing the Gower telephone before the Society of Telegraph 
Engineers (I.E.E.) in London, April 23, 1879, Mr. Scott said : 

When the user wishes to call attention at the other end of the 
line he does exactly as he would do with an ordinary speaking tube, 
with the use of which every one is familiar. He blows into the 
flexible tube, and the air pressing through the orifice in the 
diaphragm vibrates the reed, and then escapes by a vent made in 
the side of the box. 4 

Mr. Scott also referred to a method of using the telephone for 
making its own call patented by Mr. A. F. St. George of the India 
Rubber Company : 

Outside the coil of the telephone another coil of thick wire is 
wound, an intermittent current from a battery being passed through 
this outer primary coil, corresponding currents are induced in the 
inner coil, which as usual is permanently connected with the line, and 
cause vibrations of the diaphragm at the distant telephone audible 

1 U.S. specification, No. 199,007, January 8, 1878. 

2 U.S. specification. No. 217,278, July 8, 1879 (appb'cation filed October 
24, 1878). 

3 The Telephone, the Microphone, and the Phonograph, Du Moncel, 
P- 359- 

4 Journal of the Society of Telegraph Engineers, viii. 335. 


at a considerable distance. The vibrations are rendered inter- 
mittent by means of an armature vibrated by an electro-magnet. 1 

And to the method of 

Mr. Siemens, of Berlin, [who] uses a reed temporarily held in the 
mouth-piece of the telephone, by which not only is the sound 
produced by the vibration used, but it is made to act on a small 
ball which produces a series of blows upon the diaphragm or disc, 
the result being an audible signal at the further end of the line. 
This invention requires the telephone when not in use to be kept 
in an upright position. 1 

The magneto call bell had made such progress that these methods 
of utilising the telephone itself were but very little used in the 
United States, Gower's for example was never commercially used 
there, but a method analogous to that of Siemens was suggested 
by Edison in an application filed March 4, 1878. He says : 

The invention consists of a stand for the receiving instrument 
and a swinging metal lever, the end of which comes into contact with 
the diaphragm, so that it is thrown from it violently when a strong 
wave or current passes over the line or through the magnet of the 
receiving instrument. This lever, in returning, strikes the diaphragm 
a blow, and produces a sharp penetrating sound like that of a Morse 
sounder, and this may be heard in all parts of a large room. 

I have heretofore shown, as in Case No. 146, an induction coil 
in connection with a telephone. I arrange a switch between the 
local and main line circuits, in such a manner as to vary the electric 
tension on the line by moving such switch, and thereby operating 
the call at the distant station ; and I prefer to employ a peculiarly 
constructed induction apparatus, in which there is a fine wire wound 
helically round a larger wire, and then the two are wound to form 
a helix. The larger wire is in the local circuit, and the induced 
current is set up in the finer helix. 2 

This probably never went into operation. The Edison instru- 
ments were fitted with a local call bell operated through a relay by 

All these devices were merely tentative, and quite early the 
patent specifications indicate the striving for a more effective 
means of obtaining a call. The Edison interests held to the battery 
possibly by reason of the Western Union telegraphic groove. The 
same reason probably also accounts for Morse sounders being 
attached instead of bells to the earlier instruments manu- 
factured. The Bell interests, on the contrary, directed their attention 

1 Journal of the Society of Telegraph Engineers, viii. 335. 

2 U.S. specification, No. 203,017, April 30, 1878. 



to alternating current magneto machines and polarised bells, 
perhaps in the first instance in order to obtain a continuance of 
that freedom from the troubles of batteries which Bell's permanent 
magnet telephone permitted. So early as October n, 1877, we 
find Watson applying for a patent in which he frankly abandons 
one feature of the circular issued in May by saying : 

In operating magneto-electric telephones such as are described 
in the said Letters Patent, difficulty has been experienced in calling 
the attention of the operator at a distant station and it has been 
found advisable and necessary to combine with such a telephone 
instruments specially adapted for the purposes of signalling. 

FIG. 39. Watson's Magneto-electric-inductor Call. 

The object of the present invention is to produce an audible 
signal at a distant station of sufficient loudness to attract the 
attention of the operator at a considerable distance from the 
instrument. To this end I combine with a telephonic circuit a 
magneto-electric inductor of ordinary or suitable construction. 
Electrical currents of high tension may thus be induced upon the 
line by the rotation of an armature, and these currents, on being 
passed through the coils of a distant telephone, produce a sound of 
considerable loudness. I prefer, however, to combine with the 
telephones at the receiving end of a circuit a bell or other contrivance 
specially adapted for calling attention. 1 

It will be seen from the illustration (fig. 39) that Mr. Watson 
was utilising the best known form of magneto electric machine, and 

1 U.S. specification, No. 202,495, April 16, 1878. 


it may be considered of sufficient interest to reproduce an early 
illustration of such a type taken from ' The Illustrated Handbook 
of the Royal Panopticon of Science and Art ; an Institution for 
Scientific Exhibitions, and for Promoting Discoveries in Arts and 
Manufactures, 1854.' The Royal Panopticon was situated in 
Leicester Square, London, and the building which it occupied is 

FIG. 40. ' Clarke's Magnetic Electrical Machine.' 

now known as the Alhambra Theatre. Fig. 40 represents ' Clarke's 
Magnetic Electrical Machine ' there exhibited. 

Another example of the same calling device may be seen in 
Roosevelt's application of May 29, 1878. Mr. Roosevelt was 
evidently no believer in economic waste, for ' the object of [his] 
invention is to make a combination telephone, which shall consist 
of a telephone and a magneto machine, and to use a single permanent 
magnet both for the telephone and the magneto machine.' 1 

1 U.S. specification, No.' 2 18,775, August 19, 1879. 



This device was never applied to commercial uses. It was 
otherwise with the push-button type of magneto of which Anders' 
U.S. patent, No. 228,586, June 8, 1880 (applied for July 21, 1879), 
may be taken as an example. The patentee says : 

The second part of the invention consists in the combination, 
with the telephonic apparatus at each station, of a magneto 
induction apparatus operated by the depression of a push knob for 
generating the current which operates the annunciator or signalling 
apparatus at the central office. 

The illustration (fig. 41) shows the 
complete outfit and indicates that its 
introduction followed the Blake trans- 
mitter, which is enclosed in the same 
case as the magneto inductor. Though 
put into practical use it was not 
reliable, and its manufacture was dis- 
continued after one or two years. 

Another form of magneto made by 
the Western Electric Manufacturing Co. 
is illustrated in Prescott's ' Speaking 
Telephone ' of 1878, p. 24 and following 
pages, with diagrams of circuits. The 
generator was of the Clarke form with 
revolving bobbins. 

After a very short experience of such 
types as these the superiority of the 
Siemens armature pattern was recognised 
and all the makers adopted it. 

Watson, who had done so much in 
assisting Bell with his experiments leading 

FIG. 41. Anders' Push- 
button Magneto. 

to the invention of the telephone, and who devoted his atten- 
tion to developing the calling apparatus by adopting magneto 
generators as above mentioned, has also left his mark on the 
art in connection with the ' ringer ' of the magneto-bell. On 
August i, 1878, he applied for a United States patent on an 
' Improvement in polarised armatures for electric bells.' The 
improvement relates to ' that class of electric bell strikers having 
their armatures and electro-magnets polarised by proximity to, 
or contact with, the poles of a permanent magnet, and which 
are operated by alternately reversed currents of electricity.' In 
describing his apparatus he says : 

Figure [42] represents a front, and fig. [43] a side, elevation of 
my improved bell ; and fig. [44], a modification in which a single 
bell is used. 


A is one pole of the permanent magnet, and B the other. C is 
the armature, pivoted at the point d of the metal piece D, so that 
when one end is against the electro-magnet the other is away 
from it. 

E E are the supports for the bells F F. These are attached to 
the metal piece D by screws passing through slotted holes, and the 
bells can thereby be adjusted in their relation to the hammer or 
striker G. 

H H is an ordinary electro-magnet fastened to the pole B. 
Instead of the horseshoe magnet A B, one or more straight-bar 
magnets may be used, extending from the metal piece D to the 
back-piece J of the electro-magnet. A single bell may also be 
used, as in fig. [44], by attaching a hammer or striker, G G, to each 

FIG. 42. Watson's Polarized 

FIG. 43. Watson's Polarized 

end of the armature C, and fastening the bell-support to the metal 
piece D directly over the centre of the armature. The hammers 
can strike either on the inside or the outside of the bells. 

When a current passes in one direction through the coils of the 
electro-magnet, the armature is attracted at one end and. repelled 
at the ''other. When a current passes in the opposite direction this 
action" is reversed. 1 

This centrally pivoted polarised armature bell of Watson's 
at once went into general use, has survived all changes in other 
apparatus, and is still almost exclusively employed. 

The plan of using one gong suggested in fig. 44 has not been 
practically used, probably because it lacks the provision for adjust- 
ment, which is so simple and so effective with the two-gong type. 

At the first meeting of the National Telephone Exchange 
Association, representative of the Telephone Companies of the 

1 U.S. specification, No. 210,886, December 17, 1878. 



United States, in September 1880 a committee was appointed who 
examined, as carefully as the time afforded them would permit, the 
different appliances in the way of call bells for telephonic use, and 
begged leave to report as follows : 

The bells presented are from the factories of Charles Williams, 
Jr., of Boston ; Post & Co., of Cincinnati ; Davis & Watts, of 
Baltimore ; The Gilliland Manufacturing Co., of Indianapolis ; 
and the Electric Merchandising Co., of Chicago. 

In general it may be said that the bells are excellent, the 
enterprise and progressiveness of some of the manufacturers are 
most remarkable. We think the device of the Gilliland Company, 
which dispenses with the press button on the ringing of the bell, 
deserves attention. 
Also, the various im- 
provements of Post & 
Co., Charles Williams, 
Jr., and others. 

We earnestly re- 
commend to the manu- 
facturers the adoption 
of "the interchange- 
ability of parts. 1 

The design of these 
various magneto-elec- 
trie call-bells (a name 
subsequently short- 
ened by general usage 
to ' Magneto ' simply) 
varied with each maker. The Magneto contained not only the 
generator and receiving bell (termed the ' ringer ' ), but also the 
switches, cut outs, and lightning arresters; everything, in short, 
except the talking instruments and the battery. The driving gear 
was one of the important differences in each make. In the Post 
bell the pinion wheel attached to the armature was driven by means 
of a rubber band, in the Gilliland friction wheels were used, and 
in the Williams friction wheels also, but of another type. 

' The device of the Gilliland Company, which dispenses with the 
press button on the ringing of the bell,' doubtless referred to the 
automatic cut out. Since the generator coil offered considerable 
resistance a short circuiting shunt was provided and was normally 
completed, but this shunt required to be broken when a ringing 
current was to be sent out on the line. In the first instruments this 
was done by pressing a button with one hand whilst turning the 
handle with the other. Automatic devices of varying kinds were used 

1 National Telephone Exchange Association Report , 1880, p. 186. 

FIG. 44. Watson's Polarized Ringer with 
single gong. 


to overcome this difficulty. In the Gilliland a lever with a spring 
attachment was connected to the handle. The revolution of the 
latter removed the lever from a contact. In the Williams the shaft 
of the generator was normally in contact with the short circuit. 
A Y-shaped slot in a collar on the shaft altered the position of the 
latter on its revolution and thus broke the short circuit, and in 
the Post magneto a weighted spring was thrown off at a tangent 
when the armature was revolved. These were different methods 
for performing the same result and all tending towards simplifying 
the work the subscriber was called upon to do. 

FIG. 45. Phelps' Automatic Switch for pony-crown receiver. 

In the same direction was the utilisation of the receiver 
to varv the position of a hook, and thus actuate springs which 
changed the circuit so as to cut out the ringing apparatus and cut 
in the talking apparatus, or vice versa. To effect this change by the 
weight of the telephone was the subject of a patent to H. L. Roose- 
velt of New York for a device termed 'the gravity switch.' The 
existence of this patent and the royalty claimed under it led to the 
production of an alternative automatic switch adapted to the 
' pony crown ' receiver. In the United States ' pony ' is a prefix 
denoting diminutive. The crown receiver was designed with the 
idea that a number of magnets would add to the power of the tele- 
phone. The magnets were circular, and the instrument placed 
diaphragm downwards had very much the appearance of a crown. 



It was soon found that any superiority which this instrument might 
be considered to possess did not arise from its numerous magnets. 
The size was reduced and only one magnet remained. The switch 
was so arranged that the act of removing the instrument from or 
placing it upon an appropriate bracket changed sliding contacts 
and effected the same purpose as the gravity switch. Figs. 45 
and 46 show this switch as illustrated in the patent granted to 
G. M. Phelps. 1 

** The magneto was so important an adjunct of the telephone 
that much attention was given to its development, but experience 
was needed fully to impress manufacturers and telephone companies 

FIG. 46. Phelps' Automatic Switch for pony-crown receiver. 

with the need of sound design and reliable workmanship. The 
results attained during the first few years of the existence of exchange 
service were not satisfactory, as may be gathered from a paper by 
Mr. Sunny presented to the Switchboard Conference in 1887. The 
paper referred to the field for improvement in subscribers' apparatus 
generally, 2 but some of Mr. Sunny's remarks on the magneto bell 
may be more conveniently given in abstract here. 

The magneto as constructed to-day is a cheap looking affair, 
except the new Gilliland, and they are all more or less unreliable, 
while after ten years' experience we ought to have an instrument 
that would look in keeping with the furnishings of the finest 
residence or office, and that would be free from electrical defects. 

1 U.S. specification, No. 222,201, December 2, 1879 (application filed 
September 10, 1879). 

2 Chap. xxiv. p. 328. 


Chicago's experience with magnetos has been unsatisfactory, and 
this is no doubt true of other exchanges. 

He records the various types of magnetos used, and continues 

We changed to the Gilliland iron magneto, which by this time 
was a year old, and had had some of the original defects in con- 
struction eradicated. I think that the thanks of every telephone 
man are due to Mr. Gilliland for the great effort he made to give us 
a magneto that would be perfect in every respect ; and while 
electrically it has fallen short of our expectations, its appearance 
impresses the subscriber with the idea that he is getting something 
for his money, which cannot be said of the other forms of magnetos. 
The iron magneto had four spring contacts of phosphor-bronze, two 
for the local and two for the line. These contacts became so caked 
up with dirt that the trips of the repairmen were exceedingly 
frequent, and the annoyance to subscribers from being cut off, 
exasperating. In 1886 we had 2100 cases of switch trouble, chiefly 
in these magnetos. To remedy this, we cut down the number of 
contacts to three, and substituted German silver springs. This 
lessened the number of cases of trouble to 1600 in 1887 (December 
estimated) . 

Another fault with the iron magneto is the automatic cut out. 
The number of these troubles for the current year will reach 500. 
With the armature of the magneto in circuit we find in many 
instances that we cannot ring the subscriber, so that we make it a 
practice to keep the armature cut out. We detect the presence of 
an armature in circuit by testing with an apparatus that indicates 
the line open if its resistance is higher than 400 ohms. Notwith- 
standing these defects, we continue to buy the iron magnetos. They 
are heavy ; the machinery is inaccessible ; and they are constantly 
working loose from the wall because of their weight. The sub- 
scribers like them, however, and when we send a man to a new 
subscribers with a repaired Davis & Watts in a new box with shining 
bells, and all in apple-pie order, he comes back with the load on his 
back, and says that that man wants a machine like his neighbour's, 
which is a new Gilliland, or none at all. We rarely have to change 
a new Gilliland magneto for a new one, while the D. & W. are 
being crowded out of the service by the people who will not renew 
their contracts until they get a new Gilliland. 

These defects in the magneto are not peculiar to Chicago. It 
is impossible that we should be forced to take out magnetos in 
Chicago because of defects, and that the same make of instrument 
should be a complete success five hundred miles away. Yet it is a 
fact that magnetos that are undeniably defective are furnished in 
one exchange long after the defects have been proven by another 
exchange. We have three makes of instruments to select from. 
The Gilliland iron magneto, rich in construction but poor in opera- 
tion ; the Chicago magneto, and the Post. The last two gotten 
up with a view to the strictest economy. It would be better, it 


would seem, if there was but one magneto, since we cannot have 
more than three, and that that one be a carefully constructed, well 
tested instrument electrically, and in appearance something that 
will impress the subscriber as favourably as the iron magneto. 

One result of the discussion arising out of Mr. Sunny's paper 
was the use on magnetos of hinges having a spiral wire whose 
ends were soldered to the respective sides of the hinge, a device 
which materially reduced magneto ' troubles ' by preserving the 
continuity of the conductor. 

The development of the magneto call bell was carefully and 
constantly continued. The educative influence both to users 
and manufacturers was important. The ' interchangeability of 
parts ' recommended by the Association became a necessity of 
its manufacture, and the application of labour saving devices and 
improvements in factory methods which has since become so 
pronounced a feature in the production of telephonic apparatus 
generally received its first impetus in the making of magneto bells. 
Much ingenuity was applied to their design, and the manufacture 
attained a very high standard in later years ; * but at the com- 
mencement of the industry, which is the period now being dealt 
with, there were necessarily crudities both in design and manu- 
facture. Despite these drawbacks a practical instrument was 
available for the use of the pioneers in Telephone Exchange service. 
And they were prompt to avail themselves of all the facilities at 

1 Chap. xxix. p, 454. 



AT the time that the telephone was invented the telegraph had 
assumed such proportions that offices existed into which numerous 
lines entered. Some of the circuits were operated at that office ; 
others went through. The changing of the circuits was accom- 
plished by means of switches, and the switches were mounted on a 
board. Hence the name which has been retained to describe the 
most extensive and elaborate apparatus in electrical engineering. 

Although the switchboards used in telegraph offices exercised 
an influence on the development of the telephone switchboard, 
the latter served a purpose not previously effected, and the 
influence of the telegraph board was limited to a detail of construc- 
tion. Speaking generally and subject to the exceptions related in 
Chapter IX, a telegraph switchboard was not operated at the desire 
of an employee at a distant station to be connected with some other 
distant station, but by an employee at the office where the switch- 
board was placed, and usually for some purpose connected with the 
instruments therein. 

A telephone switchboard, on the contrary, is for the purpose of 
interconnecting lines at the desire of subscribers at the extremities 
of those lines. And this purpose, though not entirely new, was 
practically new to all concerned in the telephone business. Tele- 
graphic exchanges were so little used as to be generally unknown, 
and the switchboards used in connection with them were not de- 
scribed in the literature of the art. Dumont, who patented a tele- 
graphic exchange system 1 in 1851, describes in his specification 2 
a switchboard for effecting the desired results. But Dumont and 
his proposals were unknown, and telephone switchboards were 
developed on new lines to meet new needs. 

Dumont's switchboard was submitted in later years to critical 
analysis, and it was claimed that as a practical device it was inopera- 

1 Chap. ix. p. 77. * British specification, No. 13,497, 1851. 




tive because no provision was made for reversing the polarity of the 
battery, and also having regard to the type of telegraphic instrument 
with which it was designed to be used. But any minor difficulty 
of this sort would certainly have been overcome on being put into 
operation. The intention to provide intercommunication between 
all the lines is clear enough, and as the earliest known example of 
a switchboard designed for the purpose of such intercommunication 

L . - i ~mr I -- I, .^ , . I I a 

FIG. 47. Dumont's Telegraph Switchboard. Simplified diagram from 
patent of 1851. 

at the request of subscribers, the operation of the mechanism 
proposed must be briefly described. 

The drawing of the switchboard forming part of the specification 
is partly diagrammatic, but somewhat too large for identical 
reproduction. A simplified representation on a reduced scale is 
given in fig. 47. 

The parts A, B, and C in fig. 47 are reproduced in figs. 48, 
49, and 50 respectively from the patent drawing sheet 2. 

The specification has probably suffered somewhat in clearness in 
the course of its translation from the French to the English language, 
but it was apparently intended that each line entering the central 
office should pass through a two-point switch A on its way to a 


dial telegraph instrument B and earth. A call for intercommunica- 
tion having been received, the switch A was moved so as to cut 
out the instrument B and earth and put the line in connection with 
the communicators C. Fifty communicators are illustrated, and 

FIG. 48. Dumont's Telegraph Switchboard (A). 

although 200 numbers are shown on each, the whole outfit is intended 
to illustrate 100 lines only. Limiting ourselves to this 100 we will 
regard each communicator as having 100 studs around its circum- 
ference and two hands electrically connected, moving on the same 
axis at its centre. The No. I studs on all the communicators are 

FIG. 49. Dumont's Telegraph Switchboard (B). 

connected together, and so with all the other numbers. It follows 
that at any one of the fifty communicators any two subscribers' lines 
may be connected together by placing one hand on one stud repre- 
senting the calling line and the other hand upon another stud repre- 
senting the called line, the switch A of the latter having been operated 
so as to cut out the instrument B attached thereto. The provision 



of fifty communicators for 100 lines enabled all the subscribers to be 
connected at any one time, which we know now to be in excess of 
requirements. Had Dumont supplied ten communicators instead 
of fifty, he would have made provision equal to the ten pairs of 
cords per 100 subscribers generally allowed in early telephone switch- 

But this suggestion for a telegraphic exchange unhappily came 
too soon to be applied to public use or reward the patentee. 

The ' dial ' form of switch has frequently been used for diverting 
a line to one of several other lines as required. Perhaps the more 
general form is that with one hand or arm which, pivoted at one 

FIG. 50. Dumont's Telegraph Switchboard (C). 

end, makes contact at the other end with a stud, the line being 
connected with the arm. The arm revolving like the hand of 
a clock can make contact with any stud arranged at equal distances 
from the pivot centre like the figures on a clock face. To permit 
interchange amongst a number of lines, such dials may be multiplied 
according to the number of lines and the studs of each of the 
numerous dials connected together at the back of the board. 
Continuing the analogy of the clock face and assuming the 
studs to equal the hours in number, then all the i o'clocks would 
be connected together, and all the other studs respectively, up to 
twelve. Line A having its switch arm on, say, No. 3 of its dial, and 
line B having its arm on stud No. 3 of its own dial, then the circuit 
would be made up of line A through the switchboard and out at 
line B. Mr. Lockwood states in ' Practical Information for Tele- 
phonists ' (p. 79) that a switchboard of the dial type was put in 


operation by Mr. Murray Fairchild, in connection with the telegraph 
office at New Haven, Connecticut ' away back in the fifties.' 

The first telephone switchboard actually installed for regular 
business communication was situated in Chapel Street, New Haven, 

j__T . , . ' 

~L= f i!r"L2__r * 

__ J nS LJ _J p>*L J 


FIG. 51. Switchboard used at New Haven, January 1878 (circuits). 

in January 1878. The apparatus at that office is illustrated in 
the diagram, 1 fig. 51. 

Each line was connected to a lever pivoted at P, resting nor- 
mally on stud s, thence through a relay, R, and battery to ground. 
In the local circuit of the relay was an indicator drop D for each line, 
the bell of the indicator being common to all the drops. To send 
a call the subscriber opened the circuit, causing the relay to fall 
back and close the indicator circuit. On the fall of the drop the 

1 From drawing in possession of Mr. T. D. Lockwood. 

FIG. 52. Switchboard used at Meriden, Connecticut, 
February 1878 (front). 

FIG. 53. Switchboard used at Meriden, Connecticut, 
February 1878 (back). 


central office operator moved the lever of the calling line from 
its stud s to stud /, thus cutting out the relay and indicator and 
connecting the line through the stud / with the bar B. At the same 
time he turned the lever g to its left-hand stud, thus connecting the 
line with the telephone and receiving his orders. The called line 
would then have its lever diverted to its stud t, and through the 
common bar B the two lines would be connected. To send a signal 
out to line, the lever g would be turned to the right, and would 
then be in connection with a buzzer circuit. This produced a loud 
buzzing sound in the subscriber's telephone. 

The diagram shows lines i and 3 connected. 

Since there is no provision for indicating the close pf conversa- 
tion, it must be presumed that this was effected by the operator 
listening in at intervals as described in the later model. 

This switchboard was thus a combination of relays, indicators, 
and two way lever switches. It was only in use about two months, 
being then replaced by one of the dial type, which those in tele- 
graphic circles in the locality were probably familiar with through 
Mr. Fairchild's work. It would seem, however, that the dial pattern 
of board must have been used in the first instance at Meriden, 

The front and back of the Mermen board are illustrated in figs. 52 
and 53 respectively. Fig. 54 is a sketch of the connections which 
was made in after years by Ellis B. Baker, who installed the board, 
and who states that it was put into use on February i, 1878. 

The dials on this Meriden board, it will be seen, are of the one 
arm pattern, and two dials were consequently needed to be operated 
to make a connection. Simultaneous connection could thus be given 
to four out of the eight subscribers on the board. 

The operation of the board is thus described by Mr. Baker : 

The circuit started from the earth, passing through a line battery, 
pair of annunciator coils, thence to binding post marked i Out, 
thence to disk X 1 through switch lever S 1 to binding post i In, 
with a leg connecting with No. i disc of each of the circles. From 
binding post i In, the circuit passed to the subscriber's premises, 
where was located an ordinary closed circuit push button. From 
this push button the circuit passed to the ground or to the next 
subscriber on the line. Circuits 2, 3, and so forth were connected 
in a like manner to their respective posts, discs, switches, etc. 
When a subscriber on circuit i desired to obtain the attention of the 
operator, the circuit was broken by pressing the push button, 
releasing the annunciator shutter showing the number of the 
circuit. The operator threw the lever S 1 to the left against a stop, 
and the lever of circle a, which normally rested on disc 8 connected 
with the earth, to disk I, thus completing a circuit through disc i, 
lever A, telephone H, lever B, and disc 8 to earth. Upon receiving 




the subscriber's call, supposing the subscriber desired to be con- 
nected with circuit 2, switch lever S a is thrown to the right in 
connection with ground or buzz-bar, and the subscriber's number 
sounded by vibrating the handle of the buzz-box, which was 
connected between the buzz-bar and the ground. After the number 
had been repeated two or three times switch-lever B was thrown 
from disc 8 or the ground to disc 2, thus connecting the two 
circuits together through telephone H, and if the parties were in 
conversation switch-lever S 2 was thrown from the buzz-bar back 
past the disc X 2 against the stop. This completed the connection 
of two subscribers or two circuits, and the same connection could 
be made on either pair of discs. There being no way the subscriber 
could signal when the line battery was thrown off, it was necessary 
for the operator to frequently listen in to ascertain when to discon- 
nect the circuit. 1 

The Bell exchange in Chicago was started early in 1878, and no 
description is available of the switchboard first used, but about 
August of that year another switchboard was substituted. The 
following description of this switchboard is compiled and the 
illustration fig. 55 reproduced from evidence and sketches of 
employees. 2 

On a wall were placed indicators. Below and in front of them 
was a construction similar to the console of a three-manual organ. 
Upon each bank of this were placed spring jacks, not like those 
which have become so familiar, but simply long springs resting upon 
a contact. Above the console and inclined at an angle was a rack 
composed of metallic bars. 

The line came first to the spring of the jack, through the anvil 
contact to the indicator and ground. 

The metal rods were grouped in pairs, and between the rods 
of each pair was connected a clearing-out indicator. 

To connect two subscribers the operator inserted in the spring 
jack a wedge-shaped plug with a metallic top and an insulated 
base, which disconnected the indicator and ground. To the flat 
plug was attached a cord, and at the other end of this cord a gripping 
plug, which has been likened to a clothes-pin. The clothes-pin 
plug gripped one of the rods. Another cord similarly equipped 
had one end grip the corresponding rod and the other connect 
with the spring jack of the second subscriber. 

This switchboard is notable as being the first provided with 
clearing-out drops. A patent was applied for by Horace H. Eldred 
in the United States on June 9, 1880, and granted August 18, 1884 

1 Citizen's Case, Circuit Court of the United States, Western District of 
Michigan, 1896, Defendants Record) p. 213. It will be observed that the 
order of the respective parts is the reverse of the photographs of the board. 

2 Ibid., Complainants Record and Briefs, pp. 177, 201, 211. 


(No. 303,714). The first claim of this patent protects the com- 
bination of telephone lines, spring jacks ' for the insertion of con- 





necting wedges,' and ' visual signals or calling annunciators . . . 
placed in each line at a point between its spring jack and earth, 
whereby the annunciators are cut out while talking." 



The second claim covers ' an auxiliary or supplemental signalling 
apparatus included in each of said connecting conductors, whereby 
the sub-stations so connected, or either of them, may notify the 
attendant at the central station to disconnect the said lines.' 

The switchboard at New Haven, the earliest in construction and 
operation, was without any cords, whilst that at Chicago utilised 
flexible conductors between the spring jacks and connecting racks. 

FIG. 56. Western Union Peg Switch for Telegraph Offices. 

Flexible cords were not in favour with telegraph engineers, 
and were the subject of much criticism in later years amongst 
telephone engineers. But cords could be avoided without the 
use of dials, which were in fact but little used for telegraph switch- 
boards. The form generally used at the time the telephone was in- 
vented was probably that known in the United States as the Western 
Union peg switch (fig. 56) and in Europe as the Swiss Commutator 
or Universal Switch (not to be confounded with the telephone 
switch which was called ' Universal ' and will later be described). 


Though differing in form, the same principle of operation was 
employed in these switches. A series of metal bars crossed the 
board horizontally ; above them, but not in contact, was placed 
another series of bars arranged vertically. A horizontal bar could 
be placed in contact with any vertical bar by means of a metal 
plug. Another plug could connect the same horizontal bar with 
some other vertical bar. The vertical bars were continuations of 
the lines, the horizontal bars connecting-straps. 

Although the Bell licensee manufacturers (Chap. xv. p. 184) in 
the switchboards with which their names are more generally identi- 
fied followed the principles of this telegraph switchboard, Charles 
Williams, at any rate, had made other types. As early as the 
autumn of 1878 he supplied cord boards with annunciators and 
flat spring-jacks, passing connections to operators' tables where calls 
were controlled. But the types which came on the market in the 
general way as standard manufactures were those having vertical 
bars to which the lines were connected, and horizontal bars or 
' straps ' for the purpose of connecting together any two vertical 

The examples given below are taken from circulars or catalogues 
issued about 1882 or 1883, and thus do not represent the first 
productions, but are commercial forms made after three or four 
years' experience. 

The Williams switchboard is thus described in a circular issued 
by the manufacturers, the illustrations being reproduced in fascimile. 


Manufactured by 



The engraving, fig. [57], shows the present form of switchboard, 
giving a general idea of its appearance in perspective showing a 
hand telephone and Blake transmitter in position. 

It is composed of the upright board A and the inclined board 
E ; upon both are arranged the connecting strips B in series of 
four, and designated as A, B, C, D, E, F, G, H, I, and J. Arranged 
between the two boards A and E are the annunciators D. 

Fig. [58] is a sectional view of the switchboard, arranged to 
show the connections. 

In this figure four connecting strips are shown on the board A 
and four upon the board E. 

Upon the back and under sides of the board are placed the line 
strips. These are composed of metal springs arranged so as to 
normally press their free ends together as shown, and so as to 
form a continuous connection. The spring-jack R and wedge W 



shown on the front edge of board are used to loop in an operator's 
telephone and transmitter, and also for signalling a subscriber, 



FIG. 57. Williams' Switchboard. 


FIG. 58. Williams' Switchboard (section). 

the ends of flexible cord, O, N, being connected to the telephone 
and J to a magneto generator. 

M 2 


The line circuit enters at L, connecting with spring-jack R, 
through the shoe to line strip C, on the board A, a branch running 
through an annunciator D to the line strips on the table E, reaching 
earth through the ground strip /, by means of the plug P, which 
normally rests therein, and connects the strip to the line. 

To call up a subscriber, insert the wedge into the spring-jack. 
Depress the knob K, and signal with the central office generator. 

To connect two subscribers, the plugs P, P of the two lines 
which are withdrawn from the ground plate / are passed through 
the holes of a common connecting strip B, pressing between the 
springs d, of the line strip C, and making electrical contacts there- 

jooooo 10000079997 V99979V999S 

Fi'i. 59. Williams' Switchboard (skeleton diagram). 

When the operator wishes to put the listening telephone into 
circuit, the wedge W is inserted in spring- jack R. 

Fig. [59] shows a skeleton board, with the connection to tele- 
phone, generator, and ground. 

The figure illustrates a board of fifty lines. 

The boards are arranged to be placed end to end to accommodate 
any central office system. 

The price of this board is $3.00 per circuit, including lightning 
arrester, telephone and transmitter holder and patent wedge. 

The Gilliland switchboard is illustrated in fig. 60, the same general 
features being followed, but the construction of the strips and plugs 
differed from other makes, being very simple, correspondingly cheap, 
and lacking strength and durability. 

Post & Co. of Cincinnati (who later became the Standard 



Electrical Works) manufactured a board of which fig. 61 is an 

From the catalogue is taken the following description of the 


The line of posts on top of table are to connect subscribers. The 
connections for engine, telephone, microphone and night service 

FIG. 60. Gilliland Switchboard. 

bell are all properly Tagged. The night bell, No. 6, can be thrown 
out of circuit at any time desired. The operation of the table is 
as follows : A call comes in and drop No. 4 will fall, showing party 
desiring to talk. You answer by taking out plug No. I on his line 
and insert in second strip on board No. 2. Then move treadle, 
which answers subscriber by a ring. After .answering, push the 
large button No. 7, on the lower right hand corner, when you are 
thrown in circuit with microphone and telephone. After answering 
and asking what is wanted, take the same plug No. i, that you first 


used, and insert in any of the ten strips No. 3 which is not in use. 
Call up the party that is called for in the same manner above 
described. When answered by a ring, take the plug No. I on his 
line and place it in the same strip that you inserted the first plug 
in, which throws them in circuit. If at any time you desire to 
listen to subscribers conversing, push on small button No. 8, placed 
at right hand side of connection strips. 

FIG. 61. Post Switchboard. 

Trunk Lines, A, B, C, D, E, and F. Tables having trunk lines 
.11 operated in the same manner as described, with this exception : 
That when parties are wanted on any other tables, connect in strips 
marked A to F on any table desired. The 50 line table has connec- 
tions for 6 additional tables, 5 lines each. We can increase this 
number, if desired, at a very slight addition to our regular rates. 

We use no cords about the table, depending on rubbing points 
entirely, as our long experience in the matter has fully demonstrated 
the fact of that being tJte best manner of connecting. 

Subscribers should be particularly notified to ring off. Use 


two cells of Gravity Battery on the microphone circuit of switchboard, 
as it is always in circuit. The night service bell, No. 6, rings 
continuously when call comes in until drops are set by operator. 

Treadle for Generator is always off dead centre, so that the least 
motion of the foot starts it. Telephone and microphone stands have 
an universal adjustment. We furnish with each table a microphone 
month-piece, which Exchanges will find very convenient in conveying 
sound to microphone. The engine requires a drop of oil now and 
then to prevent working hard and to generate strongly. Tables 
black walnut, highly ornamented and veneered. 

The competition of rival manufacturers and the lack of capital 
on the part of the purchasers probably combined to produce a 
lightness of construction leading to cheapness in price, especially on 
the part of those manufacturers like Post & Co. and Gilliland, 
who were distant from the Bell headquarters. Williams' switch- 
boards were more solidly made. 

The switchboards already referred to are those introduced by 
the manufacturers licensed by the Bell Co. But the Western Union 
Telegraph Co. were running telephone exchanges in opposition to 
the Bell, and the manufacturing interest identified with the Western 
Union Co. was the Western Electric Manufacturing Co. The illus- 
tration of the Western Union Telegraph switch given in fig. 56 is 
from the Manufacturing Co.'s catalogue, and it is noticeable that 
whilst other telephone manufacturers took the telegraph type of 
switch for their model, the makers of that switch themselves started 
on new lines. The types of the Bell licensees were early abandoned, 
whilst that of the Western Electric Manufacturing Co. became the 
starting-point for most of the later developments. 

The relations of the Western Electric Manufacturing Co. with 
the Western Union Telegraph Co. were close. Consequently, 
when the Western Union Co. entered the telephone field the 
Manufacturing Co. became actively engaged in the production of 
telephones and accessories. The Edison transmitter was made 
by them. The receiver adopted by the Western Union and also 
made by the Manufacturing Co. was the Bell invention, pure 
and simple, though modified in form by George M. Phelps, 

But in the case of telephone switchboards, an article of new 
manufacture, the company were not indebted to any outside 
source for ideas, and even avoided following, as other makers had 
done, their own patterns of telegraphic switchboards. In Scribner's 
British patent, specification No. 4903, of November 29, 1879, is an 
illustration (fig. 62) of what was known as the ' Universal Switch,' 
from which the standard switchboard was developed. 

In describing the invention Scribner says : 


I cause each subscriber's wire to terminate in a metal block, and 
these blocks are arranged upon a board, so that there are as many 
blocks upon the board as there are subscribers. Each block carries 
a blade similar to the folding blade of a pocket knife, and, as in 
such a knife, the blade folds into a groove or recess in the block, the 
block corresponding to the handle of the knife. A spring also tends 
to keep the blade closed in its recess. 1 

Here it will be observed the knife-like block is the means of access 
to the line instead of the vertical bars of the telegraphic form of 
switch, and this knife-like block is an electro-mechanical contrivance, 
its purpose and its operation being thus described : 

In this state of affairs when the subscriber sends a call current 
through] the wire it excites the electro- 
magnet, the magnet attracts its armature, 
and allows an indicating shutter to fall ; 
this shows to the attendant that this 
subscriber calls for his attention. Each 
block (corresponding to the knife handle) 
has two holes formed through it, and 
into one of these holes the attendant 
when called sticks a metal peg which dis- 
places the knife blade, partially opening 
the knife. In this way the line wire 
is disconnected from the indicator. The 
peg inserted into the hole in the block is 
attached by a flexible conductor to a 
telephone, by the aid of which the 
attendant receives the subscriber's order 
to couple his wire with that of some other 
subscriber with whom he may desire to 
converse. Now for making these connec- 
tions a series of insulated bars are 

provided upon the lower part of the board, and the attendant, 
selecting one that is out of use, connects the caller's line wire to it 
by a flexible conductor terminating in pegs at its ends. One peg 
he sticks into a hole in the subscriber's block, and the other into 
a hole in the coupling bar. 1 

The succeeding requirement in telephone service is that so soon as 
the subscribers have finished their conversation they should be 
promptly disconnected by the operator. This requirement is met 
as follows : 

In connection with each of the coupling bars there is an indicator 
to inform the attendant when the conversation between the two 
subscribers lines, which the bar serves to couple, is complete. 1 

1 British specification, No. 4903, November 29, 1879. 

FIG. 62. Universal Switch 
(from patent). 



This specification contains further important features which 
render it a classic in switchboard patents, but in it the word ' switch- 
board ' itself does not appear. The figure illustrated is described 
as ' a front view of a coupling board arranged according to this 

The illustration fig. 62 from the 4903 1879 British patent is 

FIG. 63. Universal Switch (from Catalogue). 

diagrammatic only. The details are more clearly designated in 
fig. 63.1 

For the purpose of giving prices and dimensions this Universal 
Switch is divided into three parts : The upper portion is called 
a ' 25 Number Shutter Annunciator (12 by 13^ inches) $50 ' ; the 
middle section is ' 25 Number " Jack Knife " Switchboard (12 by 
8 inches), $20,' and the lower portion is ' 15 strap Switchboard 
(12 by io inches), $9.' The illustration shows only a 10 strap 

1 Reproduced from the Western Electric Manufacturing Co.'s catalogue 
issued in 1878 or 1879, p. 56. 


The combination, however, is called both a ' section ' and a 
' switchboard. ' 

The dimensions of each section, when mounted, are 12 inches 
in width and 32^ inches in height. The clearing-out annunciators 
and relays are set up by themselves apart from the switchboard. 1 

The ' clearing-out annunciator and relays ' are illustrated in fig. 64. 
The general description of the switchboard is as follows : 

The Universal Switch is the standard telephone exchange 
system of the Gold and Stock Telegraph Company. 

Each wire comes into the exchange through a lightning arrester 
to one of the bolts on the jack-knife switchboard (shown just below 

FIG. 64. Clearing-out Annunciators and Relays of Universal Switch. 

the annunciator in fig. [63]). From this bolt it passes to the 
corresponding annunciator magnet, and thence to the ground. 

When there is no plug in the jack-knife bolt the circuit passes 
straight through to the annunciator magnet and ground ; but 
when a plug is inserted in the hole in the bolt, this plug takes the 
line circuit, and the circuit through the annunciator to ground is 
opened. If, therefore, a plug is placed in bolt No. i, and another 
in bolt No. 2, and the two plugs connected by a cord, lines No. I 
and No. 2 are connected for talking. 

In practice the two lines are brought together by connecting 
them to the same strap on the ' Strap Switchboard ' [i.e. the lower 
part of fig. 63]. Each strap is connected to ground through a 
' Clearing-out Relay ' of 150 ohms resistance, so that the talking is 
done over this ground connection. When two subscribers are 
through talking, the one who called first signals out in the same 
way as he called. This signal works the clearing-out relay, which 

1 Catalogue, p. 57. 


closes a local battery, and drops a shutter on the 5-number 
annunciator [fig. 64]. This is a signal to disconnect at the office. 

The ' shutter annunciator ' is very delicate, and may be used 
with either a battery or a magneto call. 1 

This board, it will be observed, contained the important features 
that the insertion of a plug removed the calling indicator and earth 
(a provision also met with in Eldred's patent), but in addition 
brought the line into connection with the operator's telephone ; 
both calling indicators were out of circuit during the conversation 
and a special clearing-out indicator ' teed on ' in a branch to earth. 2 
This method of connecting a clearing-out drop was a very important 
application, whose value was to be more fully appreciated in after 
years. For some time yet the clearing-out drop was placed directly 
in the circuit, after the manner of Eldred, but in a way far 
more convenient for operating. 

The connecting bar across the board is a survival from previous 
types, but used in a different way. A cord with plugs at either end 
served to connect one subscriber's block with the connecting bar, a 
similar cord with plugs connected the same bar with another sub- 
scriber's block. Clearly, then, the only purpose of the horizontal bar 
was to complete the circuit of the two cords. That purpose could 
be served by connecting the cords together as permanent pairs. 
The point at which this connection was made afforded a suitable 
place for inserting keys by whose means the circuit might be diverted 
to telephone or generator as required. This modification is found in 
the next catalogue issued. The description and illustration are 
as follows : 


Described in this circular is believed to be the simplest and 
most expeditious in manipulation of any yet devised. The move- 
ments necessary to connect and disconnect subscribers are reduced 
to a minimum. 

Fig. [65] shows a section for fifty subscribers' wires. 

It consists of 

Fifty annunciators for subscribers' wires. 

Fifty jack-knife switches for connecting and disconnecting 
line wires and annunciators. 

Five clearing-out annunciators for signalling the discontinuance 
of communication between subscribers. 

A shelf with five pairs of keys for connecting to calling battery 

1 Catalogue, pp. 57, 58. 

2 These features are more fully described in Scribner's U.S. speci- 
fication, No. 293,198, February 5, 1884 (application filed August 23, 1879). 
See figs. 87 to 90. 


and telephone of central office and five pairs of cords and 

The annunciators 


FIG. 65. Standard Switch- 
board with ' jack-knife ' 

of a new and improved construction, 
recently perfected by this Company. 
They are much more sensitive than 
the ordinary forms. 

Their magnets are wound to a 
resistance suitable to either battery or 
magneto currents. 

The switch is furnished in sections 
of fifty wires, mounted and with con- 
nections run between the various parts, 
ready for setting up and connecting 
the line and office wires. 

The sections of switch are only 
fourteen inches wide, and can be placed 
side by side, so that one operator can 
readily attend several sections. 

Annunciators for closed circuits are 
furnished where desired. Part of the 
annunciators may be for open and part 
for closed circuit in the same section. 

All the annunciators are provided 
with local circuit points, closing a 
local bell circuit by the fall of the drop. 
This is put on all the annunciators, 
though not always used. In small ex- 
changes not requiring all the time of an 
attendant to do the switching, it is 
very useful. In larger exchanges it is 
very convenient where night service is 
required. A large vibrating bell may 
be used to awaken an attendant when 
he is required. 

By the use of the local attach- 
ments and night bell many exchanges 
have been enabled to save the wages 
of a night operator. 

The switch is intended for ex- 
changes where either magneto or battery 
calls are used ; and with our automatic 
pole changer works equally well where 
either or both systems are used in 
the same exchange, 
shows the manner in which the switch 

The diagram [fig. 66] 
is connected. 

Only five of the line annunciators and one of the clearing-out 
annunciators and one pair of keys are shown in the diagram. 

As original, but the lines should obviously read ' L 54321.' 


L : i, 2, 3, 4, 5 are the line wires. 

A i, 2, 3, 4, 5 are the line annunciators. 

J i, 2, 3, 4, 5 are the jack-knife switches. 

C is the clearing-out annunciator. 

K, is the telephone key. 

K y/ is the battery or calling key. 

P P it are plugs with flexible cords. 


FIG. 66. Standard Switchboard (circuits). 

T is the switchman's telephone. 

B is the calling battery. 

The operation of the switch is as follows : 

If annunciator No. i indicates a call the switchman inserts 
plug P, into jack-knife switch i (disconnecting line from annunciator 
i and connecting it with clearing-out annunciator C and keys K y 
and K^), and by using telephone while pressing key K, acknowledges 
No. i's call. 

He then calls the subscriber desired by No. i (say, No. 5) by 
inserting plug P,, into jack-knife switch 5 and tapping key K /r 
He then presses key K, and listens at telephone for No. 5's acknow- 


ledgment, upon receiving which he notifies No. 5 that he is 
connected with No. I, and listens until conversation begins, when 
he releases key K, and is at liberty to attend to another call. 

Subscribers are to signify the termination of their conversation 
by tapping their call keys thus dropping the shutter of clearing- 
out annunciator C. When the switchman observes this he pulls 
out the plugs from the jack-knife switches, thus restoring the lines 
to their proper annunciators and leaving the pair of cords and keys 
available for another call. 

If the subscriber neglect to signal out, the switchman can easily 

ascertain if conversation is con- 
cluded by pressing key I [KJ 
and listening at the telephone. 

For the convenience of the 
switchman the left hand or tele- 
phone key of each pair is dis- 
tinguished by a dot on the knob, 
and the cords are made in pairs, 
being of different colors. 1 

It will be noted that this 
switchboard contained the same 
number of jack-knife switches 
as annunciators, thus making the 
board complete for fifty lines and 
permitting connections to adjoin- 
ing sections, but making no 
provision for cross-connection or 
transfer lines. In the next 
commercial publication of the 
FIG. 67. standard Switchboard with manufacturers from which the 
transfer jacks on side of frame, illustration fig. 67 is taken cross- 

connection jacks are inserted in 

the frame-work at the operator's right hand. In the first quoted 
catalogue it was said that ' the Universal Switch is the standard 
telephone exchange system of the Gold and Stock Telegraph 
Company.' In the second the universal switch had developed 
into the ' Standard Switchboard,' and the name is still retained 
to indicate a simple as distinct from a multiple switchboard. The 
description given is as follows : 

This is a cord board, with the annunciator drops, and automatic 
switches and keys for calling and connecting telephone to sub- 
scribers' circuits. In this board the movements necessary to 
connect and disconnect subscribers are reduced to a minimum, 
and it is believed to be the simplest and most expeditious in 

1 Catalogue, pp. 5-6. 


manipulation of any board yet devised. It is now in practical 
operation in over two hundred telephone exchanges in this country, 
and is used on about forty thousand telephone lines. . . . 

By the use of superior stock in the construction of our cords, 
and with weights to take up the slack of cords, the trouble from 
this source is reduced to a minimum. 

Our boards are 14 inches wide. The No. i board rises 73 inches, 
and the No. 2 board rises 60 inches from the floor. 

In the details of construction of its various parts this board 
was to undergo numerous changes, but in its arrangement of calling 
indicators, jacks, and clearing-out indicators on the upright, with a 
horizontal keyboard for manipulating keys it reached at once the 
most convenient form, and in these essential features has remained 
unaltered. In the next model produced after fig. 67 the transfer 
jacks were placed below the line jacks and in the same panel. 

An important change in detail will be noticed between the 
knife-like blocks in fig. 66 and the fig. 67 illustration. The 
jack-knife switch is superseded by the ' spring jack.' The oval 
form of the front of the jack in fig. 67 is due to the large shoulder 
with which the jack was then furnished, being fixed to the board 
from the front with two screws. In. the next model the shoulder 
was abandoned and numerous illustrations of it have been published. 

Electrically they are the same, the difference is in form, because 
it was early realised that space on the front of the board was valuable, 
hence the mechanism was constructed to operate from front to 
back instead of from side to side. 

It has sometimes been assumed that the term ' spring jack ' has 
been derived from a spring form of jack-knife switch. But this is 
not so. In the telegraph switchboard a spring pressing on a contact 
was described as a ' spring jack,' and the name was carried on from 
telegraphic to telephonic apparatus. 

Whilst referring to questions of terminology it may be remarked 
that the signalling device was called an annunciator in the United 
States, but in Great Britain ' indicator ' was more general. The 
difference reflects the previous practice in the use of electric bells. 
In the United States a device for signalling the number of the 
room from which a call had been transmitted was an electric bell 
' annunciator,' whilst in Great Britain it was an ' indicator.' For 
telephonic purposes the ' Indicator ' gradually became general in 
both the Old World and the New. They were also known as ' shutter 
drops,' which became shortened to ' drops.' This was especially 
the case with the indicators for signalling the close of a conversation, 
which became known as ' clearing-out drops.' 

The development of a switchboard on new lines is an indication 
that the experts of the Manufacturing Co, recognised at an early 


stage that the problem of telephonic switching was a new problem 
and needed to be solved on different lines from that of telegraphic 
switching. In telegraphy the switching was an incident, the con- 
trivance by which it was effected was only in occasional use. 
Facility in the operation of the mechanism and the time employed 
by such operations were of subsidiary importance. In telephony 
it was otherwise, and whilst this was eventually brought home to 
everybody, the Western Electric Manufacturing Co. must be credited 
with realising it well in advance of the lessons of actual experience. 
This is evidenced by the production of the Standard Switchboard, 
which represents not merely a collection of parts admirably adapted 
to serve their respective purposes, but also an arrangement of those 
parts in such a manner as to afford the greatest advantage in use. 

The arrangement of the indicators and jacks was such as to show 
readily their respective relationships. A decimal arrangement was 
adopted, the reading being from above to below. The clearing-out 
drops had an obvious relationship to the cords with which they 
were in circuit, and the keys permitting the insertion of the generator 
or speaking instruments were not only original in the work they 
performed, but were placed upon a shelf at the most convenient 
position for the operators. For the first time also the equipment 
of the board was the result of study. The number of subscribers to 
which one operator could in general attend was found to be such 
that fifty was taken as the most convenient standard, the number 
of cords was determined after inquiry as to the probable maximum 
of simultaneous connections, and on the same principle twenty 
per cent, was adopted. 

The framework of the switchboard was a striking departure 
from the elaborate ornamentation and overlapping cabinet work 
of its contemporaries. The simplicity, however, cannot be con- 
sidered as a tribute to the aesthetic principle that surplus orna- 
mentation is reprehensible. It was due to the recognition of the 
economic value of frontage space on switchboards. Ornament 
which increased the frontage space was something which not only 
cost money at the start, but, what was still more important, might 
entail an unnecessary expenditure in operators' wages and in rent. 

The framework was consequently so arranged that a new board 
could be added, fitting closely to the existing board or boards so 
that an operator was not limited in connections to her own board, 
but could complete a connection on either of the adjoining boards, 
thus reducing the number of ' cross-connections ' or transfer.^. 

Some of the parts and circuits of the standard board are the 
subject of patents to Scribner, Kellogg, and Warner, but the 
design of the board is believed to be largely due to F. R. Welles, 
who appreciated the importance of the decimal principle, the readily 



observable relationship of the indicators with their allied jacks, 
the facilities which the keyboard afforded, and the utility of a 
compact constiuction permitting to the operators the common 
use of adjacent sections. 

In Prescott's ' Speaking Telephone ' of I884 1 the Western Electric 
Standard Switchboard is illustrated and described 2 as the ' Gilliland 
Switchboard.' On p. 
288 it is called ' one of 
Gilliland's Switch- 
boards,' and the real 
Gilliland Board is illus- 
trated on the same 
page, and called the 
' Gilliland Standard 
Switchboard.' In 
Preece and Maier's 
' The Telephone'' 
(1889), p. 314, an illus- 
tration of the Western 
Electric Standard 
Board is described as 
the Williams Switch- 
board. Whilst many 
manufacturers de- 
scribed their switch- 
boards as ' standard,' 
the word has acquired 
a meaning as a self- 
contained switchboard 
as distinct from a mul- 
tiple board, and it is 
the form illustrated in 
figs. 65 and 67 which 
is so understood. 

An example of a 
number of Williams switchboards treated as independent units is 
seen in fig. 68, illustrating the central office at Portland, Oregon, re- 
produced from the New York Electrical Review of April 17, 1884. 

All the switchboards above described were constructed on the 
principle of providing complete equipment for an operator who 
received a call, connected talking instruments into the line to ascer- 
tain the number required, and plugged the two subscribers together 
when their positions on the switchboard permitted. If the called 
subscriber's position on the switchboard was beyond the reach of 

FIG. 69. Switchboard used in Coleman 
Street, London, 1879. 

1 P. 277. 

2 P. 276. 


the operator who received the call, it was necessary for her to 
obtain the assistance of a second operator. This was usually effected 
by means of transfer lines, to which the term ' trunks ' or ' office 
trunks ' was applied. The first operator would connect the calling 
subscriber's line to a ' trunk ' and desire the second operator to 
connect the same trunk to the called subscriber. Thus there was a 
division of labour between two operators in some cases, but not in 
all. There were, however, at an early date, switchboards in which 
a division of labour was provided for on an organised plan, whereby 
each did some portion of the work, and all calls were completed by 
the intervention of more than one operator. Such methods have 
been more developed in recent years, as, for instance, in ' distributing ' 
systems. An early example of this division of labour in switching 
may be seen in the illustration fig. 69, in which one operator does 
the work of connecting and disconnecting, another the talking. This 
board was used in the Bell Exchange in London in 1879, and is 
believed to have been of Williams' s manufacture. 

Here the calling line was connected by the switchman to the 
table, and the operator at the table received from the subscriber the 
information as to the number required, instructing the switchman 
accordingly to make the necessary connections. 



THE first public circular issued in May 1877 l was signed by Hub- 
bard and Watson, who were acting on behalf of Bell and the 
others interested as well as of themselves. Bell was the inventor, 
Hubbard and Sanders were financial supporters, whilst Watson was 
a technical assistant sharing with Bell the hopes and fears of 
experiments, and carrying out, with the aid which Williams's 
workshop afforded, the construction of apparatus that Bell 

In a few months after the issue of the first circular the develop- 
ments were such as to show the need of regularising the ownership 
and preparing for expansion. Consequently in August (1877) the 
Bell Telephone Association was formed but not incorporated. This 
Association consisted of Bell, Hubbard, Sanders, and Watson. The 
patents were transferred to the Association. The shares of Bell, 
Hubbard, and Sanders were equal three-tenths each ; the remaining 
tenth was allotted to \Vatson. During October 1877 a contract 
was entered into by Hubbard ' as trustee of said patents,' but in 
November contracts were made by the Bell Telephone Company 
(i.e. the Association above mentioned) and signed by Hubbard as 
trustee and Sanders as treasurer. 

The New England Telephone Company was incorporated on 
February 12, 1878, with a capital of $200,000, and was granted the 
exclusive right to use, license others to use, and to manufacture 
telephones in the New England States. The Bell Telephone Com- 
pany was also incorporated on July 30, 1878, with a capital of $45,000, 
with the object of extending the use of the telephone throughout 
the United States outside qf New England. These two companies 
were subsequently amalgamated under the name of the National 
Bell Telephone Company, which was incorporated March 13, 1879, 
with a capital of $850,000. In consequence the first New England 

1 Pp. 67-68. 

179 N 2 


Telephone Company went out of existence, but the name was 
revived later when a company was formed under that title to 
amalgamate a number of smaller original licensees and to carry on 
the exchange business throughout that territory. 

The policy of the holders of the patents was to appoint agents 
in denned localities authorised to lease telephones to the users 
thereof at certain stated rentals, of which rentals the agent should 
retain a specified share. In a contract dated October 24, 1877, there 
is no mention of exchange systems, though the reservation to the Bell 
interests of the exclusive right to make contracts with ' any and all 
parties who desire to use telephones for the purpose of transmitting 
messages for hire ' probably contemplates the district system. 

The term ' telephone ' is defined as meaning instruments made 
under the Bell patents, Nos. 161,739,174,465, 178,399, and 186,787, 
and all patents which Bell had or thereafter might obtain for 
improvements, and all beneficial modifications which he should be 
at any time authorised to use. The term ' one set ' is apparently 
defined rather in reference to the payments contracted for than to 
the practical working, since it includes either ' four small telephones, 
two large telephones, or one large and two small telephones.' 

The agent or licensee undertakes to construct with his own capital 
' in the most approved way, any and all lines which may be reason- 
ably required ' in the area specified ' by any proper person, for use 
in connection with telephones. Said lines to be hired or sold on 
reasonable terms. The rental price may be a gross sum, as rental 
for telephones and line, in which event the schedule rates herein 
shall be the sum apportioned as the rental of telephones.' 

The licensee further undertakes to use his ' best efforts in all 
proper ways to introduce telephones to the utmost possible extent, 
and to procure lessees thereof for use in ' the defined territory. He 
undertakes to ' employ at least one suitable and efficient agent whose 
whole time shall be devoted to the introduction and care of tele- 
phones, and will employ such further and other agents and instru- 
ments as may be necessary for said business.' The contract from 
which these quotations are made covered an entire State, so that ' at 
least one ' does not seem at this date to be an extravagant require- 
ment, but the phrase is illuminating as illustrating the caution of one 
of the contracting parties whatever may have been the expectations 
of the other. It may be remarked parenthetically that according to 
the official census the number of persons employed in the telephone 
business in the same State on December 31, 1902, was 2823 and 
would now be much more. 

On November 15, 1877, the Bell Telephone Company issued 
' Instructions to Agents. No. i ' which commences with the state- 
ment that 


In consequence of the difficulties that have arisen in different 
localities for want of uniformity in price for the rental of telephones, 
the Bell Telephone Company has adopted the following rates for 
all its agencies, and prices are to be fixed in accordance herewith. 

The annual rental for telephones shall be ten dollars each, 
payable in advance ; not less than a pair of telephones must be 
used at each station, except as hereafter specified. 

For social purposes, single telephones may be used at each 
station. By ' social purposes ' is meant the use of telephones 
as a matter of convenience between private houses ; between a 
house and private stable ; a doctor's house and office, etc. etc. 

For district telephone purposes a discount of twenty per cent., 
and for house use a discount of fifty per cent, may be made, and 
the use of single telephones allowed at each station. 

By ' house use ' is meant all places where telephones are used 
in one building, or group of buildings, as, for instance, several 
buildings in the same yard used by the party ; or, in fact, w r here 
telephones substantially take the place of speaking tubes. College 
lines may be included in this line. 

The magneto bell calls may be sold for fifteen dollars each, 
or rented for five dollars each per annum. 

On March 8, 1878, the Bell Telephone Company entered into an 
agreement with the District Telegraph Company of St. Louis, who 
agreed ' to introduce the Bell telephone as a part of its district 
system, to substitute telephones for the district boxes as rapidly 
as possible, to construct private lines of telephones, and to prosecute 
the introduction and rental of telephones with all due diligence.' 

On May 31, 1878, a contract was entered into with the Connecticut 
District Telephone Company whereby the District Company was 
given the exclusive license for the term of ten years to use telephones 
' for district purposes ' in New Haven and certain other cities. 

The District Company was also licensed ' to connect said cities 
as above named by wire, and to transmit messages between them 
for hire by means of the Bell telephone, provided that such con- 
nection is made within one year from the date of this contract ; 
but this license shall not be construed to be an exclusive license for 
such purposes.' In the first year 500 telephones were to be leased. 

On July 3, 1878, an agreement was made between the Bell 
Telephone Company of New York and the Bell Telephone Company 
of Boston whereby the latter gave the exclusive right ' to use and 
rent telephones ' at the annual rental of $10 ' for general purposes ' 
and $5 ' for house use.' 

The New York Company agreed that its capital stock should 
consist of 1200 shares of preferred stock and 800 shares of common 
stock, each of the par value of $50. The capital was to be furnished 
by the holders of the preferred stock. The ordinary stock was 


paid to the Bell Telephone Company of Boston as a consideration 
for the exclusive right. 

The territorial system is provided for in the following clause 
the ' party of the first part ' being the New York Company and 
' the party of the second part ' the Boston or parent company : 

The party of the first part agrees that whenever the party of 
the second part shall be prepared to receive business in any of the 
towns or cities within said district for transmission to places outside 
of said district, the said party of the first part will turn the same 
over to the party of the second part, it paying a reasonable com- 
pensation for receiving or collecting same, and also turning over 
to the party of the first part the delivery of messages received 
by the party of the second part from points outside of said district 
to be delivered within said district at a reasonable compensation. 

Nothing in this agreement shall be construed to prevent the 
party of the second part from establishing offices within said district 
for the transmission of messages to points outside of said district. 
And no consolidation, sale or change, or insolvency or dissolution 
of said party of the second part shall affect the rights or privileges 
granted to the party of the first part by these presents. 

A contract entered into on January 29, 1879, for the first time 
defines the ' district use ' in the following terms : 

It is further understood and agreed and made a part of the 
above and within agreement, that the terms ' district and exchange 
purposes ' and ' district uses/ mentioned in the above and within 
agreements, refer to the use of telephones in connection with a 
district system to be established within the territory covered by 
this license, more particularly described as follows : 

A central office or receiving station is established within the 
prescribed limits, from which point lines of wires are erected running 
in different directions within the prescribed territory. At some 
of the principal factories, stores, shops, offices, business places, 
dwellings, etc., along any of these said lines of wires, telephones are 
placed and so connected in the line that persons at any of such 
connected points in the line, after the proper adjustment of switches, 
cut-outs, instruments, etc., and exchange of proper signals, can 
communicate with each other on the same line, and with the said 
central office or receiving station, with parties located on other of 
said lines of wires having a similar telephone connection. 

A further development in the definition of ' district or exchange 
purposes ' is reached in a contract of singular clearness and 
definiteness in all its clauses. It is dated August 9, 1879, and 
appointed C. H. Haskins agent for the States of Wisconsin and 
Minnesota, and is signed Theo. N. Vail as General Manager of the 
National Bell Telephone Company. 


The terms ' District ' and ' Exchange ' purposes as used in this 
agreement are understood to apply to a telephone business in which 
in any city or town one or more telephone circuits are established 
and connected with a general office, or general offices, to receive or 
execute orders or to establish connections between different lines. 

The district business includes the right to transmit messages for 
hire over such lines, but does not include the right to transmit 
messages for hire between different cities or towns. 

The reservation of the long distance system is also shown in 
this agreement by the following clause : 

This Company [i.e. the Bell Company] reserves the exclusive right 
of renting telephones for the purpose of communication between 
different towns or cities or of transmitting messages for hire, and 
of renting telephones to corporations or individuals whose business 
may be but partially carried on within the territory assigned to you, 
although one or more of said telephones may be used within the 
said territory. 

The policy of appointing agents for given areas, of supplying 
telephones on lease, and of reserving for the company the com- 
munications between distant places was acted upon by Hubbard ; 
and Bell, in his letter to the financial supporters in London, 1 
emphasised the desirability of preventing sale outright, but the 
progress by the early part of 1878 was so great and the promise 
so much greater that Hubbard realised the importance of placing 
the organisation of the business in strong hands. From his own 
experience in the control thus far he realised to the full the need 
of active and far-seeing management. In July 1878 Mr. Vail became 
general manager, retiring from an important position in the United 
States Postal Service for that purpose. 

Capital, hitherto supplied by Hubbard and Sanders, was now 
needed in quantities far beyond their slender resources. Some 
citizens of Boston were sufficiently impressed with the work 
accomplished and the future possibilities, to invest in the Bell 
Company. They appointed as President Col. William H. Forbes, 
a man of influence and strength of character, who left his impress 
upon the enterprise. The company was now fairly embarked upon 
an active career under auspices which thus provided influence and 

The country was mapped out, agents appointed, and agreements 
entered into providing for the development of a business of which 
all were ignorant because it had never existed before. 

The essential feature of this business was the telephone, and 

1 P. 89. 


provision was made for its manufacture by Charles Williams. 
But it was found that call bells were also necessary, and as time went 
on switchboards were required. Very naturally the Bell personnel 
took an active interest in the development of such accessory 
apparatus, and the company adopted the policy of licensing 
manufacturers in various parts of the country. From these manu- 
facturers the exchange licensees were authorised to purchase such 
accessory apparatus as they might need. 

The following were the manufacturers so licensed : 

Charles Williams, Jr., Boston. 

Post & Co., Cincinnati. 

Gilliland Electric Manufacturing Co., Indianapolis. 

G. H. Bliss & Co., Chicago. 

Davis & Watts, Baltimore. 

Quite early in its career, therefore, the Bell Company exercised 
an influence in the use of apparatus outside the telephone itself, 
and sought such methods of control as should ensure that its 
licensees obtained apparatus of approved quality, whilst leaving 
to such licensees the selection of any of the types furnished by 
those various manufacturers. 

The developments by 1880 were such as to need the formation 
of yet another company with a still larger capital, and the American 
Bell Telephone Company, which was organised in May of that 
year, succeeded the National Bell Company. 

The first report of the American Bell Company, issued in March 
1 88 1, shows that the general lines of its policy were to grant licenses 
for local use, to retain in its own hands the long distance service, 
and to study all the practical questions involved for its own benefit, 
and the advantage of its licensees generally. The following extracts 
from this first report to the shareholders relate to these various 
features : 

The policy of making only five year contracts was adopted, in 
order that our company could have time to learn the best permanent 
basis for the relations between the company and its licensees, and 
to see which of them would prove satisfactory as associates. Many 
applications are now being made for permanent licenses, and we 
have begun to give such permanent contracts in places where 
the business is being prosecuted with energy and success in exchange 
for a substantial interest in the stock of the local companies. By 
pursuing this plan, the company will gradually acquire a large 
permanent interest in the telephone business throughout the 
country, so that you will not be dependent upon royalties for a 
revenue when the patents shall have expired. 

The business of connecting cities and towns by telephone wires 
has been taken up in the past year with some vigor, and the prospect 


is good for a large increase in these lines. Boston, for example, is 
now^n communication with seventy-five cities and towns, including 
Providence, Worcester, Springfield, Lawrence, Lowell, and other 
important places. 

It will take some time yet to get first rate service in a large net- 
work of towns, as the practical difficulties at least equal those which 
were met in giving prompt connection within the limits of one city ; 
but nothing but experience and tests of various methods are needed 
to enable such groups of exchanges to reach satisfactory results. 

A large amount of work has been done in the electrical and 
experimental department, both in examining new inventions and 
testing telephones and apparatus, and in studying the question of 
overhead and underground cables, and the improvement of tele- 
phones and lines, for both short and long distance service. This 
work is expensive, but it is of the first importance to our company, 
and must be continued. 

The report of the General Inspector upon this department shows 
that we own or control, either by purchase or by inventions made 
by our own electricians, 124 patents, and have applications to the 
Patent Office for 77 more. Among these a considerable number 
are of great value as a protection to our business, and from them 
a substantial revenue has already been received by royalties from 
our licensees. This source of income will be materially increased, 
and should eventually more than cover our experimental and 
electrical expenses. 

The report concludes with an expression of the directors' 

appreciation of the ability, fidelity, and zeal with which the general 
manager and his assistants have grappled with the unusually per- 
plexing difficulties encountered in systematising our affairs. 1 

In after years the general manager, Mr. Vail, himself related 
in a brief preliminary statement of a legal nature, the principles 
underlying the organisation of the company, and remarked that 

The American Bell Telephone Company early foresaw the 
possibilities which were offered by the development of the inven- 
tions covered by these telephone patents, and that it would be 
possible to establish a great system of intercommunication through- 
out the whole country. 

The ideal of the Company would not have been realised unless 
the users of the telephone in the territory of each of the associated 
companies had been enabled to communicate with the users of the 
telephone in the territory of every other company. 2 

1 American Bell Telephone Company's Report, 1881. 

z Statement and Answers to Interrogatories. Missouri, July 27, 1910. 



FOR the purpose of illustrating the principles upon which the 
industry was organised, the last chapter includes the period 1877 
to the formation of the American Bell Telephone Company in 1880. 
But this period was so momentous in its events that it must again 
be referred to. 

Brief reference has already been made to the Western Union 
competition. Strong in influence, with ample financial resources, 
with lines extending throughout the country, with agents every- 
where, and employing a large number of men skilled in electrical 
work, it is easy to conceive the confidence with which the Western 
Union Company set out to oppo -e the Bell. It is less easy at this 
distance of time to realise and do justice tc the resolute manage- 
ment of the Bell Company, whose material resources were slender, 
but whose faith in their patents was unshaken, and whose deter- 
mination to develop the business was promptly and continuously 

The saving of the situation by the Bell Company may be 
attributed to the dual factors of prosecution of patent rights and 
the creation of Bell exchanges wherever possible, whether in com- 
petition with Western Union exchanges or not. Apart from the 
fact that a business concern like the Western Union would attach 
importance to the business results of competition, it must be 
remembered that where any doubt exists, or may be alleged to 
exist, as to the originality of an invention, a patentee who has only 
a patent occupies a much less strong position in protecting that 
patent than one who, in addition to his patent, has a business 
created out of it. A patent is granted for reasons of public utility, 
and the existence of a business is a proof of that utility ; but a 
patentee who is content to hold a patent without energetically 
working it, or arranging for its being worked, does not confer 
upon the public the benefits for which prima facie the patent was 
granted. The practical creation of a business on another's inven- 



tion gives the creator no rights in equity, and in no way 
diminishes the infringement in law. The danger to the inventor 
is a possible bias in the interpretation of doubtful or arguable 

It is not suggested that such ideas were in the minds of the 
Bell management, so far as the contemplation of the eventual results 
was concerned. The evidence available clearly shows that they 
realised the enormous public benefit which must follow from the 
use of the telephone, and especially from the exchange system, 
with corresponding financial benefit to those who controlled it. 
They realised also that Bell was entitled to the credit for making 
telephonic communication possible, and they believed that Bell's 
patents gave them the control. That they were undaunted in 
engaging in litigation with a foe much better equipped in material 
resources was doubtless largely due to their conviction of the 
strength of their case and their recognition that, even unarmed, 
justice is a considerable resource in conflict. 

From amongst the various places in which the Western Union 
interests had established exchanges, Boston was selected to form 
the subject of a test case. The suit was technically against a 
Western Union agent named Dowd. The real defendants in the 
case were four : The Western Union, The American Speaking 
Telephone Company, the Gold and Stock Telegraph Company, 
and the Harmonic Telegraph Company. 1 

The Harmonic Telegraph Company was owned or controlled 
by Elisha Gray and S. S. White, ' a very rich man in Philadelphia 
who established dental depots all over the United States.' J The 
American Speaking Telephone Company was organised to develop 
the Western Union telephone interests. Gray with his partner 
held one-third interest in it. 

Testimony was prepared, but the case never came to trial. 
Experts in electrical science and in law had examined all the 
evidence that could be got together, and advised the Western Union 
that it was impossible to plead anticipation or to impeach the 
validity of the Bell patents. 

Mr. Frank L. Pope, a well-known electrical and patent expert, 
advised to that effect, and Mr. George Gifford, an eminent lawyer 
who was leading counsel for the Western Union, reached the same 
conclusion. In consequence, overtures were made with a view to 
a settlement on terms. The nature of these overtures and the 
result were subsequently recorded by Mr. Gifford in an affidavit, 
in the course of which he stated that in the years 1878-1879 he was 
one of the counsel for the Western Union Telegraph Company. 

1 Dickerson's Argument, New Orleans Case, p. 123. 


At that time the Gold and Stock Telegraph Company, a company 
connected with the Western Union, had manufactured and were 
controlling the use of many thousands of telephones (there were 
about 10,000), and had established telephone exchanges, auxiliary 
to their telegraph business, in the city of New York and elsewhere. 
The telephones controlled by the company were composed of a 
receiver, generally known as the magneto receiver, and under- 
stood to be substantially the thing described in Bell's patent ; 
but they were of a form which was claimed and had been constructed 
by Phelps and Gray. The transmitters were carbon microphone 
transmitters, constructed under the plan of Edison and Phelps, 
and contained the induction coil covered by the Page patent, also 
owned or controlled by the Western Union Telegraph Company. 
Among other defences set up, the European publications relating 
to Reis's invention were relied upon, and it was alleged that Bell's 
telephone, as described in his patent, was not capable of talking. 
Elisha Gray was set up as a prior inventor, and the inventions of 
Edison and Dolbear were pleaded. A very vigorous defence was 
(he says) made by the Western Union Company, and testimony 
at great length and great expense was taken in support of the 
answer. After the testimony was closed, or substantially closed 
on both sides, he (Mr. Gifford) was convinced that Bell was the 
first inventor of the telephone, and that the defendant Dowd had 
infringed Bell's patent by the use of telephones in which carbon 
transmitters and microphones were elements, and that none 
of the defences set up could prevail ; Mr. Gifford advised the 
Western Union Company to that effect, and that the best policy 
for them was to make some settlement with the complainants. 
For the purpose of effecting such a settlement the position of the 
Western Union Company was (Mr. Gifford contended) very strong. 
In addition to the patents of Edison, Gray, and others, they owned 
or controlled what was known as the Page patent, which covered 
the induction coil used in the transmitters and was of great 
importance to them. Under his advice (he continues) a negotiation 
was opened with the Bell interests. He met Mr. Chauncey Smith, 
counsel for the Bell Company, by arrangement at the White 
Mountains, where they remained for a week in negotiation. He 
opened the negotiations on his part by admitting that Bell's patent 
was valid, and that the defendants infringed it, but he claimed 
that all the patents should be put together and that the Western 
Union should have one-half interest. This was refused and 
the negotiations failed. They were renewed by the principals 
in New York and resulted in the surrender by the Western Union 
Company to the Bell Telephone Company. Instead of the one- 
half interest claimed, the Western Union accepted one-fifth, and 


they agreed to go out of the business, transferring their exchanges 
at cost price. 1 

The one-fifth interest referred to was not the proportion of 
exchange receipts, but ' 20 per cent, of all rentals or royalties actually 
received or rated as paid in accordance with the provisions of this 
contract from licenses or leases for speaking telephones (exclusive 
of call bells, batteries, wires, and other appliances, or services 
furnished or performed) ' (article i). The American Bell Telephone 
Company did not sell instruments to operating companies, but 
supplied them under specific license for their use, and the Western 
Union were entitled to one-fifth of the amounts received in respect 
of the instruments supplied. The full text of the agreement was 
published by the Electrical Engineer (N.Y.) of August 28, 1895. 2 

Thus ended the first and one of the most formidable attacks on 
the Bell patents and business. It was yet too early perhaps to 
realise that the great advantage conferred by combining existing 
exchanges instead of continuing them in competition was not limited 
to the companies concerned, but that there was a public advantage 
as well. The underlying principle had yet to be demonstrated in 
after years. 

The Bell Company was now able to devote its energies with 
redoubled vigour to the actual prosecution of the business, with the 
result that the directors were in 1881 able to report : 

The number of our instruments in the hands of our licensees 
in the United States : 

February 20, 1880, was 60,873, 

February 20, 1881, was 132,692, 

showing an output for the year of 71,819 instruments. This 
number includes 20,885 taken over from the Gold and Stock Tele- 
graph Company [Western Union] and in use by our licensees. 3 

It is now usual to take a ' station ' as the unit for exchange 
statistics. A ' station ' requires both a transmitter and a receiver. 
It was customary in earlier years for the American Bell Company 
to record both transmitters and receivers as instruments. At the 
period in question it is probable that a number of magneto telephones 
were still in use as transmitters. Whilst the exact type of instru- 
ment, or the exact number of stations cannot be determined, it 
may be assumed that then, as now, two instruments composed 
a station, and therefore to facilitate subsequent comparisons 

1 Dickerson's Argument in American Bell Telephone Co. v. Overland Tele- 
phone Co., Circuit Court of the United States, District of New Jersey, 1884, 
P. 65, etc. 

2 Vol. xx. p. 208. 

3 American Bell Telephone Company's Report, March 29, 1881, p. i. 


the above figures are transcribed in terms of ' stations ' as 
follows : 

February 20, 1880, 30,436 stations, 
February 20, 1881, 66,346 

the latter including 10,442 stations taken over from the Western 

The agreement with the Western Union brought within the Bell 
Company's sphere of influence another manufacturing organisation, 
which had been engaged in electrical work for many years, whose 
facilities were considerable, and whose experts had shown, as 
already indicated, high appreciation of the requirements of exchange 
service. The Edison or Western Union business was absorbed by 
the Bell Company, whose personnel and influence dominated the 
situation, except as regards the manufacturing interest which was 
destined to absorb the principal licensee manufacturers of the 
Bell Company, and to take an important part in subsequent 



SCIENTIFIC men the world over evinced an immediate interest 
in Bell's invention, and some of them were as unstinting as Sir 
William Thomson in their praise of the inventor's achievement. 
Public attention was also early centred on the possibilities which 
might arise from it. 

The French alone among Governments made recognition of a 
national character. Napoleon expressed his appreciation of the 
value of scientific investigation and artistic development in 
saying that 

the sciences which honour the human understanding, the arts which 
embellish life, and transmit great actions to posterity, ought to be 
specially patronised by an independent Government. 1 

And in 1802, when First Consul, he founded the Volta prize ' as an 
encouragement to him who by his experiments and discoveries, 
shall make in electricity and galvanism a step comparable to that 
which has been made in those sciences by Franklin and Volta.' 
This prize, originally of the value of 60,000 francs, altered to 50,000 
francs when revived by Napoleon III, and so continued under 
the Republic, was conferred upon Bell in recognition of his invention 
of the telephone. The first award of the Volta prize was to Ruhm- 
korff in 1864 for the induction coil, the next to Bell, who was also 
created an officer of the Legion of Honour. The University of 
Heidelberg conferred upon him the honorary degree of Doctor of 
Medicine, in recognition of the use of the telephone in surgery. 
In 1902 the Society of Arts in London awarded him the Albert 
medal, and in 1913 the Royal Society conferred upon him the 
Hughes medal. 2 

1 Napoleon's Political Aphorisms, p. 44. 

2 A bequest was made to the Royal Society by the late Prof. David 
Edward Hughes, the ' income from which was to be annually awarded, either 
in money or in the form of a medal, or partly one and partly the other, for 



But whilst the scientific value of the telephone was highly com- 
mended, and the public had vague ideas of great potentialities, 
its commercial utility was not fully recognised, and there were 
even some hints that it might turn out to be merely a scientific 

The commercial developments in Europe were initiated by the 
American patentees with the help of a few influential and far-seeing 
people in each country. The importance which Bell attached to 
securing foreign patents, and the delay which took place in de- 
positing the British application, resulted, as we have already seen, 1 
in the United States application being deposited at a much later 
date than otherwise would have been the case. 

Some efforts to deal with the foreign patents were made before 
the development of the exchange system in the United States, but 
the efforts were promptly renewed, and with greater prospects of 
success, when the great public utility of that system had been 
demonstrated. This demonstration was effectually shown during 
the competition between the Edison interests (represented by the 
Western Union Company) and the Bell Company. The competition 
was extended to foreign fields, the alliance which was effected in 
America not applying to the foreign organisations. The intro- 
duction of the telephone in Europe was therefore undertaken with 
the additional energy resulting from the rival claims of Edison and 
Bell, as well as some others of local origin. 

Coming later into the field, Edison had more widely protected 
his carbon transmitter and was active in the exploitation of patents, 
as the manufacturing company allied with his interests was also 
active in its attempts to dispose of its apparatus. 

Special companies were formed by the owners of the Bell patents 
for the development of foreign enterprises, and so soon as the 
National Bell Company was established and under capable manage- 
ment, Hubbard himself visited Europe in order to promote the 
foreign business. 

The International Bell Telephone Company was formed in New 
York for the purpose of introducing the telephone exchange service 
on the continent of Europe, and the Tropical American Telephone 
Company to develop the business in South America, Central 
America, and the West Indies. 

Colonel Reynolds of Providence, Rhode Island, came to London 

the reward of original discovery in the physical sciences, particularly elec- 
tricity and magnetism, or their applications.' The Hughes medal was 
awarded to Bell in 1913 'on the ground of his share in the invention of 
the Telephone and more especially the construction of the Telephone 
Receiver* (Year Book of the Royal Society, 1914, p. 175). The terms of the 
award are not very happily expressed, but comment thereon is reserved 
for Chapter xxxiii. 
1 Chap. v. p. 45. 


to dispose of the Bell patents, and succeeded in interesting some 
important financiers, who formed ' The Telephone Company,' of 
which Mr. James Brand, an influential merchant, was chairman. 

A circular, dated May 24, 1879, was issued by this Company. 
It contains a number of illustrations as ' a few examples of the 
applications of telephones for practical purposes.' They are all 
of the domestic or private line order, and the circular may be 
regarded as intended to develop that branch, but reference is made 
to the exchange system indirectly : 

The development of the telephone in England, although it has not 
made such rapid strides as in America, has, since its introduction, 
been advancing slowly and surely. A large number of instruments 
are now in constant use, and it has been found that the more 
accustomed one becomes to the telephone, the more its advantages 
are appreciated. But in the former country, by means of the 
central system, one communicates with one's tradesmen, calls 
cabs, transacts business of all kinds, without going out of the room. 1 

Edison's representative in London was Colonel Gouraud, at that 
time the resident director of the Mercantile Trust Company of 
New York. He formed the Edison Telephone Company of London, 
of which the Right Hon. E. P. Bouverie was chairman. 

Both these companies started exchanges in London about the 
same time the autumn of 1879. As exchanges had been growing 
in the United States since early in 1878, it has sometimes been 
suggested that Great Britain was somewhat dilatory in taking 
advantage of telephone exchange facilities. The reason, however, 
is to be found in the position of the respective companies as regards 
the patent situation. The Bell magneto telephone by itself was 
not powerful enough for general use as an exchange instrument, and 
The Telephone Company (Bell) could not use the Edison transmitter, 
whilst the Edison Company could not use the magneto receiver. 

The first public exhibition of the Edison carbon transmitter 
in England was at the London Institution in a lecture by Professor 

1 The prices given in this circular (for a copy of which I am indebted to 
Mr. J. W. Ullett) are as follows : 



* d. 

Telephone, Ebonite, hand pattern . . i i o ] Subject to royalty 

Telephone, Snuffbox pattern . . o 15 o > of IDS. 6d. per 

Telephone, Wood Box pattern . . i 10 o J annum each. 

(The Royalty may be commuted at any time 

by a payment of five years in advance.) 
Telephone, Call Bell with Push and Automatic 

Switch . . . . . . .300 

[The policy of sale was subsequently modified and only leasing allowed.] 




Barrett on December 30, 1878. A few weeks earlier one of the first 
long-distance experiments was made between London and Norwich 
over the private telegraph line of Messrs. Colman. In all these 
experiments a magneto receiver was used. The Bell Company drew 

v^v Tl ** ^ *^*^^<^ O"<^ *A4^ ^V^-Cu 

FIG. 70. Edison's Letter to Gouraud on Loud-speaking Receiver. 

attention to this fact, and intimated their intention of taking pro- 
ceedings to prevent the infringement of their patent. The carbon 
transmitter without a receiver was useless, and the commercial 
adoption of the magneto receiver would be sure to involve litigation. 
The position of affairs was reported to Edison, and he forth- 
with produced a receiver on an entirely different principle. The 


apparatus was not destined to be in use long, because it was less 

well adapted to the 
requirements of ser- 
vice than the mag- 
neto receiver, but 
its opportune pro- 
duction permitted 
the Edison Tele- 
phone Company in 
London to com- 
mence business free 
from the liability of 

FIG. 71. Electro-motograph Principle. interference by the 

Bell Company. 

The use of this receiver was confined, in the main, if not entirely 
to England. Bell's United 
States patent controlled the 
transmission of undulatory 
currents corresponding to 
the aerial vibrations pro- 
duced by speech. The 
British and other foreign 
patents were of less scope, 
and consequently, whilst 
Edison's new receiver was 
not used in the United 
States, it was a very valu- 
able acquisition to the 
Edison Telephone Company 
of London. 

The letter in which 
Edison advised Colonel 
Gouraud of the despatch of 
the new instruments was 
reproduced in facsimile by 
means of another of Edison's 
inventions the electric pen. 
Fig. 70 is photographed 
from one of the copies. 

One reason for the haste 
with which these instru- 
ments were dispatched was FIG. 72. Edison Loud-speaking 
that they might be used by Receiver. First form produced. 
Professor Tyndall in his 
forthcoming lectures on the subject of modern acoustics. They 

O 2 


were first exhibited in operation at No. 6 Lombard Street, on 
March 14, 1879, and on the iyth were described in The Times. 
More detailed descriptions appeared in Nature of March 20, and 
Engineering of March 21. The illustrations figs. 71, 72, and 73 are 
from the last mentioned. 

In designing this instrument Edison utilised a principle which 
he had discovered some years previously. In his U.S. patent, 
No. 221,957, dated November 25, I879, 1 he says : 

The peculiar action upon which this invention is based was 
patented by me January 19, 1875, and numbered 158,787. An 
application of this action to telephony was also applied for by me 

July 20, 1877, No. 141, in which 
there is a band of paper moving 
beneath a point connected to the 
diaphragm. This feature therefore 
is not broadly claimed herein. 
The present application consists 
more particularly in devices which 
make the invention perfectly prac- 
ticable for use in commerce and 
render the same reliable and 

The instrument which was pro- 
duced at such short notice was thus 
rather a development than a new 
invention. Nearly two years pre- 
viously he had applied for protection 
on the application of the ' peculiar 
action ' to telephony, though only 
to a musical telephone with a make- 

and-break contact analogous to that of Reis. 2 Pressed for an 
independent receiver, Edison promptly set to work to complete 
the instrument in a practical form. 

The principle illustrated in fig. 71 may be briefly described : 
A stylus connected with one pole of a battery being drawn over 
a strip of paper laid upon a metallic surface, Edison found to be 
subject to the effect of friction when the key forming part of the 
circuit was open and to be free from the effect of friction when the 
key was closed. He applied this principle of obtaining movement at 
a distance as an alternative to the armature and spring of a tele- 
graphic instrument. It was useful as an invention and a patent 
rather than in practice, so that the electro-motograph, as he called 
it, was not generally known. When a telephone receiver on a new 

1 Application filed March 31, 1879. 

* Telegraphic Journal, vi. 383, September 15, 1878, quoting L'lectricitt. 

FIG. 73. Edison Loud- 
speaking Receiver (principle). 


principle was required, Edison's mind naturally reverted to the 
electro-motograph. It was originally invented to take the place 
of an electro-magnet and its armature. It was required now to 
take the place of an electro-magnet and its disc armature. Some 
audacity must have been required to assume that the variations 
in the friction would follow so closely the minute variations of a 
telephonic current. But a diaphragm with an arm attached to 

FIG. 74. Edison Loud-speaking Receiver (commercial model). 

its centre, a revolvable chalk cylinder upon which the arm pressed, 
and an electric circuit being ' thrown together ' soon demonstrated 
that the instrument was a practical telephone receiver. 

In its commercial form the instrument was reduced in size 
and attached to an arm projecting from the transmitter so as to 
be opposite the speaker's ear, as shown in fig. 74. 

Edison's new receiver spoke loudly. This tended to increase 
popular wonderment, and was assumed to be a virtue. It formed 
one of the principal claims to the attention of capitalists. A 
' private and confidential ' Memorandum entitled ' Edison's 
Loud Speaking Telephone ' commenced with the statement that 


The telephone which Mr. Edison has perfected in the course 
of this year differs radically from all previous instruments of the 
kind. Professor Graham Bell's invention, hitherto the most widely 
used, includes a magnet and a coil, and the sound is transmitted 
along the wire, losing much of its force on the way. In Mr. Edison's 
instrument the voice is mechanically reproduced at the end of the 
wire, and the speaker is heard with a volume of tone and a distinct- 
ness equal to the original utterance. 

This circular was issued in the summer of 1879, and the next 
paragraph indicates the surprise felt in London at the growth then 
effected in telephone exchanges on the other side of the Atlantic : 

The extent to which telephony has developed in the United 
States in the course of the last few months is almost inconceivable. 
In towns as large as Chicago or Philadelphia, or as small as Wilming- 
ton, what are known as telephone exchanges have been started, and 
have been taken up on a large scale. 

A line for demonstration purposes was erected between No. 6 
Lombard Street (Colonel Gouraud's office) and the office of the 
Equitable Insurance Company of the United States, in Princes 
Street. Nine other lines were added and connected with a switch- 
board at No. 6 Lombard Street. Though necessarily of an experi- 
mental character, since there was no capacity for growth in the 
central office, this was the first exchange in London on the Edison 
system. It was in use for some time previously, but was publicly 
opened in September 1879. On the 6th of that month The Times 
described the system, remarking that 

telephonic intercommunication on a practical working scale has at 
length become an accomplished fact in the City of London. . . . 
The stations, or more properly speaking the private offices, which are 
connected with the exchange are situated No. I in Copthall 
Buildings [Messrs. Parrish], No. 2 in Old Broad Street [Pullman 
Car Association], No. 3 in Suffolk Lane [Messrs. Renshaw], No. 4 
in Lombard Street [Colonel Gouraud], No. 5 in Princes Street 
[Equitable Insurance Company of United States], No. 6 in Carey 
Street, Lincoln's Inn [Messrs. Waterhousel, No. 7 in Queen Victoria 
Street (the offices of the Company), No. 8 in George Yard, Lombard 
Street [Messrs. Kingsbury], No. 9 in Throgmorton Street [Messrs. 
Anderson], No. 10 being our own establishment. 1 

In the course of a leading article on the subject The Times 
remarks : 

We publish in another column the extraordinary new uses for 
1 The Times, September 6, 1879. 


which this invention has been found capable. . . . There is no limit 
whatever to the number of points between which communication 
can be established, and scarcely a moment's delay in bringing them 
into connection with one another. ... At the present moment 
there are ten favoured spots at which this privilege can be obtained ; 
but there may just as easily be ten hundred or ten thousand, and, 
doubtless, before long there will be. 1 

The exchange was transferred to No. n Queen Victoria Street, 
and subscribers connected as quickly as possible. A list dated 
February 20, 1880, contains 172 names. In the provinces great 
activity was shown in starting exchanges both by the Bell and 
Edison companies. 

The ' Telephone Company ' working under the Bell patents 
issued a circular 2 which is not dated, but is believed to have been 
issued in September 1879 (a second edition is dated November 10, 
1879). It states that 

A telephonic exchange has been established in the city. 

Each subscriber has a wire from his own residence or office, with 
the necessary instruments attached, to the Telephone Company's 
office. A signal from the subscriber is answered by the clerk in the 
central office, who instantly makes a connection with the wire of 
any other person with whom communication is desired, and con- 
versation can then be carried on with ease and privacy without 
the possibility of any third person hearing what is said. 

In the third edition of the circular, dated December 24, 1879, 
' A telephonic exchange ' has developed into the plural ' Telephone 
exchanges have been established* in the city.' About 200 names 
of subscribers are given in this circular. The Bell Company's 
first exchange and offices were at No. 36 Coleman Street, E.G. 3 

Retaining for itself the control of exchanges, the Bell Company 
on September 2, 1879, 8 ave a restricted license to Messrs. Scott 
& Wollaston for the use of telephones for private lines and 
domestic purposes. -Gower made a slight modification in the 
Bell instrument which was alleged to be of great advantage. Gower 
acquired Scott & Wollaston's license and formed the Gower- 
Bell Telephone Company. A circular issued by this Company 
commences with the statement : 

There are, broadly speaking, four classes of telephones : 

1. The Original Telephone of Professor Bell. 

2. The Electro -Chemical Telephone of Mr. Edison. 

1 The Times, September 6, 1879. 

2 For which I am indebted to Mr. J. W. Ullett. 

3 The illustration on page 343 of Brault's Histoire de la Tttephonie shows 
this building with derrick on roof. 


3. The Gower-Bell Telephone, formed on the principle of the 
Bell Telephone, but much more effective. 

4. The Gower-Bell Loud Speaking Telephone, the latest and 
best form of instrument, being the combination of a special form 
of the Microphone Transmitter of Professor Hughes, F.R.S., with 
the Gower-Bell Telephone as a receiver. 

and makes the following criticisms of its rivals : 

The Bell Telephone, the original and beautiful invention of 
Professor Graham Bell, and the foundation of the telephone system, 
produces very weak sounds in comparison with this company's 
special form of instrument, and for this reason its commercial use 
cannot be recommended. 

The Edison Electro-Chemical Telephone can scarcely be con- 
sidered a practical instrument. Its use has been altogether aban- 
doned in the United States and on the Continent, and the authorities 
in this country find that they cannot make it work satisfactorily. 

The Gower-Bell Company was subsequently absorbed by the 
Consolidated Telephone Construction and Maintenance Company, 
a manufacturing company which, seeking new markets, introduced 
a further competition of English origin in some foreign countries. 
Under the auspices of this company there was formed the River 
Plate Telephone and Electric Light Company, with an exchange 
in Buenos Aires, and in combination with the Edison interests 
exchanges were established in Vienna, Lisbon, and Oporto. 

The competition between the Bell and Edison interests in the 
United States was closed by reason of the broad patent which Bell 
obtained there, an agreement having been come to because the 
Western Union Company acknowledged the breadth and validity 
of the original Bell patent, which really covered all practicable 
methods of transmitting speech electrically. The British patents 
were not so wide. Bell controlled the magneto telephone, except 
with a membrane diaphragm which was free of Bell's patent by 
reason of prior publication in the English Mechanic, and Edison 
controlled the carbon transmitter. The Edison Company, having 
carried on their introductory experimental work with a magneto 
receiver, became independent of the Bell patents on the production 
of the electro-motograph. The Bell Company, on the other hand, 
were incurring some risk of attack from the Edison Company by 
the use of the Blake transmitter. But no definite attack was 
made on the ground of patents. Commencing their exchange work 
in earnest about the same time (September 1879), they carried 
on their respective enterprises with all the energy which rivalry 
produces, and speedily demonstrated that they were supplying 
a service of great public utility. A common foe is a material aid 


to alliances, and to the British telephone companies in the year 
1880 the common foe was the Government. The circumstances 
under which the Post Office claimed to control telephone exchanges 
and the result of their action thereon will be recorded in a later 
chapter, but it may be remarked here that the action taken by the 
Government facilitated the combination of the Bell and Edison 
interests, and no longer necessitated the separation of the carbon 
transmitter and magneto receiver, which would appear to be the 
natural complement of each other. For much as we may admire 
the inventive faculty which produced the electro-motograph receiver 
at the moment it was required, much as we may recognise the 
commercial ingenuity with which its loud speaking features were 
commended to public consideration, it is obvious that as a practical 
working instrument it was far inferior to the Bell form. To operate 
it at all a crank had to be continuously turned by hand. The 
loudness, in ordinary exchange use, was not merely of no value 
but was a definite drawback. That this was recognised by the 
Edison interests it would be incorrect to say, but the adoption 
of the Bell receiver was recommended to the directors of the United 
Company by the engineer of the Edison Company (E. H. Johnson) 
in a report dated June 3, 1880, on account of the greater simplicity 
in operating the apparatus. The following is an extract from this 
report : 


This is a question of vital importance, for the reason that the 
telephone, unlike all other apparatus for communication at a 
distance, such as the various systems of telegraphs, is primarily 
so simple as to require no skill whatever in its use. 

Now since the general public is notoriously incapable of grasp- 
ing the simplest mechanical operations, this simplicity must be 
preserved. Any added complications of mechanism limits dispro- 
portionately the number of people who are able to manipulate the 

It is idle to seek to ' Educate the Public ' to the observance of 
' Rules and Regulations.' Such cannot be enforced except against 

In view of these facts it is, in my opinion, an essential feature in 
effecting the necessary signalling, switching, etc., etc., that these 
operations should be reduced to the purely Automatic Action of 
taking the Telephone in the hand and laying it down (or, more 
accurately speaking, hanging it up). No lack of intelligence, for- 
getfulness, or imperfect comprehension of rules can operate to 
prevent the performance of so simple an act. 

In order to obtain this degree of simplicity I have been compelled 
to sacrifice the superior qualities of the Electro-Motograph Receiver 
of Mr. Edison to the superior fitness of the Magneto Receiver of 


Professor Bell. This will appear more evident from a citation of a 
few of the operations performed automatically by the movement of 
a hook from which the Hand Telephone is taken when used, and 
upon which it is placed when the instrument is not in use. 

Mr. Johnson's introductory observations are worthy of note. 
They record an early appreciation of the work which may suitably 
be put upon the subscriber, but his reference to the ' superior 
qualities ' of the electro-motograph receiver and the ' superior 
fitness ' as a detached instrument of the magneto receiver is less 
acceptable. The loyalty of the Edison entourage to their chief 
has been recorded by Mr. George Bernard Shaw, who for a time 
was employed by the Edison Telephone Company of London, 
being described in a list of the staff as ' Way leave Manager.' 1 
Mr. Johnson's phraseology may be excused as a retreat under cover, 
but then, as now, the Bell receiver had the ' superior qualities ' 
as well as the ' superior fitness.' 

The prospectus of the United Telephone Company was issued 
on June 8, 1880, with a capital of 500,000, of which 200,000 in 
shares was allotted to the Bell Company, and 115,000 in shares 
to the Edison Company. In the course of the prospectus it was 
said : 

The telephone system in this country has not hitherto been 
properly developed, partly in consequence of the antagonistic posi- 
tion of the Bell and Edison Companies. A similar state of things 
existed in America until these two interests became united. It is 
therefore to be expected that the telephone system will now make 
rapid strides in this country, as it has already done in America. 

It is right to mention the contention of the Post Office, that their 
monopoly under the Telegraph Acts extends to the Telephone 
Exchange System, and a suit is now pending to have that question 

1 ' You must not suppose, because I am a man of letters, that I never 
tried to earn an honest living. I began trying to commit that sin against 
my nature when I was fifteen, and persevered from youthful timidity and 
diffidence until I was twenty-three. My last attempt was in 1879, when a 
company was formed in London to exploit an ingenious invention by Mr. 
Thomas Alva Edison a much too ingenious invention as it proved, being 
nothing less than a telephone of such stentorian efficiency that it bellowed 
your most private communications all over the house instead of whispering 
them with some sort of discretion. This was not what the British stock- 
broker wanted, so the company was soon merged in the National [United] 
Telephone Company, after making a place for itself in the history of literature 
quite unintentionally, by providing me with a job. Whilst the Edison 
Telephone Company lasted it crowded the basement of a huge pile of offices 
in Queen Victoria Street with American artificers. . . . They adored Mr. 
Edison as the greatest man of all time in every possible department of science, 
art, and philosophy, and execrated Mr. Graham Bell, the inventor of the rival 
telephone, as his satanic adversary ; but each of them had (or pretended to 
have) , on the brink of completion, an improvement on the telephone, usually 
a new transmitter.' (The Irrational Knot, by George Bernard Shaw, preface, 
pp. ix-x. London : Archibald Constable & Co., Ltd., 1905.) 


settled. The directors are advised that that contention will not 
succeed. Thejprivate telephone business of the company could not 
be touched under the Post Office Acts, and that department must 
pay for the right to use for profit the instruments protected by the 
Company's patents. 

The United Telephone Company itself worked the London 
system and formed subsidiary companies to work provincial 
exchanges under license. 

In France the telephone was, at an early stage in its history, 
the subject of much interest to scientific men. Du Moncel was 
one of the first in Europe to write a treatise upon it. This was 
translated into English and was for a time the standard work. 
Though published in 1879, it contains no reference to the exchange 
system. At a later stage when exchanges and trunk lines were 
making rapid progress it is to French literature that we have also 
to look for their earliest record. Wietlisbach in Switzerland had 
written upon Industrial Telephony, but mainly with a technical 
application. Brault in Paris about the same period (1888) published 
his ' Histoire de la Telephonic et Exploitation des Telephones en 
France et a 1'Etranger,' in which the industrial features had the 
greater prominence. His work stood alone then in its record of 
world progress and is of much value for reference now. On the 
practical side Ader made a modification in the Bell instrument, and 
whether by reason of the modification or from any improvement 
in manufacture his receiver was much appreciated in European 
markets. He also produced a transmitter following closely the 
carbon pencil form of Hughes' microphone, which was one 
of the best of that kind. But whilst scientists devoted attention 
to the telephone as an instrument, in France, as elsewhere in 
Europe, the exchange system was developed under American 

The rivalry of Bell and Edison was extended to Paris, but 
Gower had to some extent forestalled them, and received the first 
concession which applied to Paris, Lyons, Marseilles, Bordeaux, 
Lille, and Nantes. The Bell interests received a concession for 
Paris only, whilst the Edison interests were so fortunate as to cover 
the same cities as Gower. All these concessions were combined 
into one ownership in the Societe Generate des Telephones on 
December 10, 1880, and (except Paris and Lyons) only after that 
date were the exchanges proceeded with. 

In Germany interest in the telephone was enhanced by the 
prior work of Reis. Patriotism is not a reliable aid to the investiga- 
tion of scientific claims, but though unfruitful and forgotten, 
Reis had made experiments and constructed apparatus which he 


called a telephone, and these were sufficient to enable patriotic 
Germans to feel a proprietary interest in the invention. 

On October 4, 1877, von Stephan, Director of Telegraphs, wrote 
to Bismarck that he had established communication between his 
office in Berlin and the suburb of Friederichsburg. He intimated 
also that he contemplated at once a practical application of 
the new invention in the Imperial Telegraph Service, proposing 
the connection by telephone of country post offices to which the 
telegraph service had not yet been extended. By the end of 1877 
fifteen villages had been so connected with the general telegraph 
system. The exchange system in Berlin was not inaugurated 
until April i, 1881. 

The first exchange in Switzerland was that at Zurich, operated 
under a concession granted to a group of business men associated 
with the International Bell Telephone Company on July 24, 1880. 
During 1881 exchanges were opened in Geneva, Lausanne, and 
Winterthur by the Government, who also shortly after bought 
up the Zurich exchange. Fourteen exchanges were in operation 
at the end of 1883, and twice as many a year later. 

The first experiments in Belgium were made in 1878. 
A company was formed in Brussels in 1879, and others followed. 
Competition was recognised as unsatisfactory, and the various 
companies were encouraged to amalgamate. The Compagnie Beige 
du Telephone Bell was formed in 1882, this company being the 
Belgian subsidiary of the International Bell Telephone Company 
of New York. 1 

A similar company was formed in Holland under the name of 
the Nederlandsche Bell Telefoon Maatschappij in 1881. 

In Austria the first exchange was established in Vienna in 
1881 by the Vienna Private Telephone Company, but the 
exploitation was continued by a company under the auspices 
of the Edison-Gower Bell combination represented by the Con- 
solidated Telephone Construction and Maintenance Company of 

In Italy the International Bell Telephone Company established 
exchanges in Milan, Turin, and Genoa, and exchanges in a dozen of 
the other largest cities were started in 1881 by other interests under 
the auspices of a group of Paris financiers. 

The International Bell Telephone Company was also re- 
sponsible for the introduction of the telephone into Russia, 
Norway, and Sweden. In 1880 franchises were secured for 

1 Belgium was the locale of the first European factory of the American 
telephonic organisation, the International Bell Telephone Company and the 
Western Electric Company combining to form the Bell Telephone Manu- 
facturing Company at Antwerp in 1882. 


Christiania and Drammen, and in 1881 exchanges were established 
by the International Bell Company in Stockholm, Gothenberg, 
and Malmo, but in St. Petersburg (or Petrograd) and Moscow not 
until 1883. 

Efforts to introduce the telephone into Spain were made by 
various interests until the Spanish concession became almost a 
byword amongst concession hunters. In 1885 exchanges were 
opened in Madrid, Barcelona, and Valencia. 

In Portugal a concession was obtained and exchanges started 
by the Anglo-Portuguese Telephone Company under the auspices 
of the Edison-Go wer Bell interests previously referred to. The 
Lisbon exchange was opened on July 2, 1881, and that in Oporto 
in 1883. 

In India the development was undertaken by the Oriental 
Telephone Company, exchanges being opened in January 1882 
at Calcutta, Rangoon, Madras, Bombay, and Colombo. 

The Telephone Company of Egypt (a subsidiary of the 
Oriental Company) established exchanges in Cairo and Alexandria 
in 1880. 

Amongst other early exchanges may be mentioned Honolulu 
(1880), Rio de Janeiro (1881), and Valetta, Malta (1883). 

In the Argentine Republic there were three sources of tele- 
phonic enterprise. One of local origin with a Belgian instrument 
known as the Pan-Telephone was introduced by Mr. Fels, another 
by the Tropical American Telephone Company, and a third by the 
River Plate Telephone and Electric Light Company, formed by 
the Consolidated Telephone Construction and Maintenance Com- 
pany. The exchanges of the local Pan-Telephone Company 
and the Tropical American Company were combined under 
the name of the United River Plate Telephone Company, and 
this took over later the Consolidated Company's interests, forming 
the United Telephone Company of the River Plate, which now 
exists as one of the important foreign companies under British 

The introduction of the telephone into Australia was made 
under the Edison patents. Early in 1880 Mr. F. R. Welles of 
the Western Electric Manufacturing Company left New York for 
Australia, and in conjunction with a local firm Messrs. Masters & 
Draper established the Melbourne Telephone Exchange Company 
which made rapid progress. Efforts to obtain concessions in 
the other Australasian Colonies were less successful, but the 
attention of the local Governments having been drawn to the 
advantages of the exchange system, they shortly after established 
exchanges of their own, as did also the Government of New 


The Melbourne Exchange and those subsequently established 
at Ballarat and Sandhurst, having been purchased by the Govern- 
ment, were taken over by them on September 22, 1887. There were 
then 1019 subscribers to the company's exchanges, and of these 
752 were at Melbourne. ' The Victorian Year Book 1887-8,' l from 
which these figures are obtained, states that ' An exchange has also 
been opened at Geelong, on a guarantee that not less than forty 
persons become subscribers.' 



THE various companies formed for establishing telephone exchanges 
in Europe and abroad based their expectations of success on the 
results attained in the United States. They were able to point to 
accomplished facts indicating the great utility of such a service. 

In the United States the public generally were extremely scep- 
tical of its general adoption, and even those who could foresee the 
great public advantage were not disposed to let their hopes run too 
high. The district telegraph system was an important aid to 
development in that it permitted demonstrations of interchangeable 
connections amongst a few subscribers with but little work or 
outlay. The lines were there with call boxes connected up. To 
attach telephone instruments was a comparatively simple matter. 
In Europe there were no district telegraph systems, the telephone 
exchange was adopted in its complete form, and consequently 
expensive construction work was needed in order to permit of the 
demonstration of the utility. But the pioneers on both sides of 
the Atlantic were undaunted, and provided the capital which was 
required to tide their enterprises over the initial stages and embark 
them upon highly successful careers. 

The public apathy in the United States was as pronounced as 
in Europe. Appreciation came sooner, but only as a result of 
demonstration. Various causes have been suggested for the earlier 
adoption and more rapid development in the United States. With- 
out examining these in detail or contesting any of the numerous 
arguments which have been based on them, a sufficient explanation 
may possibly be found in the confidence and commercial energy of 
the exploiters, together with the economic conditions prevailing. 
The great distances separating American cities and the consequent 
delay incidental to postal communication contributed to the general 
use of the telegraph for communications which would have been 
forwarded through the post in Europe. It is doubtful also if at the 
time of the introduction of the telephone the local telegraphic 



service was either so cheap or so generally used in the United States 
as in Great Britain. Thus the comparatively local telephone 
(as it was upon its introduction) supplemented the long distance 
telegraph and superseded the strictly local district messenger 
service. The high cost of labour in the United States in 
comparison with Europe naturally tended to the more ready 
adoption of any expedient which economised labour or time. The 
cheapness of labour and also the cheapness of the telegraph 
service were urged as reasons why the telephone exchange system 
would be likely to be less successful in Great Britain than in the 
United States. 

In 1879 a Select Committee was appointed by the House of 
Commons to consider some questions relating to electric lighting. 
On May 2, Sir William Preece was asked by Lord Lindsay : 

As to the question of induction, you spoke of the use of the tele- 
phone in all this research that you have been making on this subject ; 
that is, of course, an extremely delicate instrument for testing ; but 
do you consider that the telephone will be an instrument of the future 
which will be largely adopted by the public ? I think not. 

It will not take the same position in this country as it has already 
done in America ? I fancy that the descriptions we get of its use 
in America are a little exaggerated ; but there are conditions in 
America which necessitate the use of instruments of this kind more 
than here. Here we have a superabundance of messengers, errand 
boys, and things of that kind. In America they are wanted, and 
one of the most striking things to an Englishman there is to see how 
the Americans have adopted in their houses call bells and telegraphs 
and telephones, and all kinds of aids to their domestic arrangements, 
which have been forced upon them by necessity. 1 

Lord Lindsay's question was interpolated in an inquiry on 
another subject, and Sir William Preece's answer may therefore 
be considered as not having been carefully weighed, but it certainly 
repeats in very similar terms the remarks made by him a few days 
earlier (April 23) in a discussion at the Society of Telegraph Engineers 
on a paper by Mr. Scott on ' Recent Improvements in Professor/"' 
Bell's Telephones.' The ' improvements ' were those of Gower. 
Mr. Scott had complained of the backwardness in the adoption of 
the telephone in England in comparison with the United States. 
Regarding this, Sir William Preece said : 

The telephone had been used in this country to a large extent, but 
there does not appear to be the want of it in England that there is 
in America. One thing which strikes one in America is the enor- 
mous extent to which they apply the telegraph and the telephone 
for their own domestic purposes. 

1 Report Lighting by Electricity, June 1879, p. 69. 


In Chicago, where there are from 7000 to 8000 calls daily, there is 
scarcely a house which has not in its hall a call bell, by which you 
may dispatch a message ;for^a doctor, or a porter, or anything else 
you want, and the reason they are driven to that is necessity 
being the mother of invention that it acts as a substitute for 
servants. Here we have no difficulty in getting servants if we pay 
them, but the difficulty in America is to get ' buttons ' at any price 
to run about for you as in England, and the result is the absence of 
servants has to a certain extent compelled the Americans to adopt 
this system of telegraphy for their own domestic purposes, and the 
telephone is to be found in almost every house as the only available 
substitute for the old system. 

Few have worked at the telephone much more than I have. 
I have one in my office, but more for show, as I do not use it because 
I do not want it. If I want to send a message to another room, I use 
a sounder or employ a boy to take it ; and I have no doubt that 
is the case with many others, and that probably is the reason why 
the telephone has not been more adopted here. The efficiency of 
the instrument in England has been seriously interfered with by 
those fearful inductive effects which are not felt to the same extent 
in America, because they have no long underground lines and they 
do not use that fast speed apparatus which produces such a tre- 
mendous roar with us. It is impossible with sixty wires in an under- 
ground pipe to speak through the telephone, the inductive effects 
being so great. 1 

These observations may therefore be taken as the expression 
of a then prevalent idea in England, an idea which also presumably 
governed official action or inaction. 

Few men have done more than the late Sir William Preece 
to popularise the telephone, or have given expression to greater 
appreciation of its beauties as a scientific instrument. It was his 
pride that he had brought the first pair of Bell's perfected type to 
England. At the Royal Society, the British Association, and other 
meetings he had lectured on the telephone, and the researches 
which he and his associates had made upon or with it. To him 
Hughes communicated his first ideas on the microphone. Sir 
William Preece may well be considered to have been an enthusiast 
on the telephone, and he was electrician to the Department which 
controlled the telegraphs. The Department refused to purchase 
Bell's patent, and the Department has been charged with short- 
sightedness and English people generally with backwardness. But 
in thus refusing, the Government who controlled the telegraphs in 
England were acting in no wise different from the company which 
controlled the greater part of the telegraphs in the United States, 
though the opinion of the latter had altered before 1879. 

1 Journal of the Society of Telegraph Engineers (I.E.E.), viii. 337. 


The failure in both cases to anticipate the actual results may be 
ascribed to a too long acquaintance with the word ' message.' 
Telegraphists sent messages, and they were keen on sending them 
as quickly as possible. It was to send messages to villages that von 
Stephan first saw utility in the telephone, but except for such 
purposes the telephone would not be likely to appeal to a 
telegraphist as an ideal means of transmission. There were instru- 
ments in existence which would send more words in a minute and 
would in addition leave a record for reference. 

The saving in time which would result from having the message 
delivered direct to the receiver might have been expected to be 
apparent, but the time element in the delivery of messages had been 
carefully studied, and it was probably considered that this saving 
would not appeal to the public sufficiently to justify the expectation 
that they would in large numbers pay for the expense of erecting 

The real difference between the people who foresaw great results 
for the telephone and those who were disposed to weigh comparisons 
of costs and differences of local conditions lay in the word ' message.' 
The telephone made practicable what had been considered im- 
possible before distant conversation. Question and answer imme- 
diate, spontaneous ; the meanings of words accentuated by inflexion 
and emphasis. The ' message ' was no comparison for this. As 
well compare the deaf and dumb alphabet with full, free, eloquent 
speech. The telephone brought distant people together with all 
the advantages of close converse which audibility permitted. Such 
advantages were not to be weighed up against the cost of messengers, 
though time showed that even so the telephone effected an economy 
as well as a revolution. 

Not from the light of after events comes an expression of surprise 
that anyone who took part in a telephonic conversation could 
have any doubts of the telephone's commercial success. To have 
conversed through the telephone before it became generally known 
was an experience which must be regarded as a privilege. The 
prevailing feeling was that of awe and wonderment, tempered by 
unbelief. When the circumstances left no room for doubt that 
the speaker was really at a distance, the clear transmission of his 
words, the proof by prompt reply that he had heard equally clearly 
the words you yourself had uttered, produced an overwhelming 
impression that here was an instrument fated to be of enormous 
value to the public, and that should produce commensurate profit 
to its introducers. 

Public as well as official apathy gave place to appreciation 
wherever exchanges were established. Commencing in the United 
States in 1878, ' in towns as large as Chicago or Philadelphia or as 


small as Wilmington ' as the ' Memorandum ' prospectus of the 
London Edison Company explained extending to London in 1879, 
by 1 88 1 or 1882 telephone exchanges were established in the 
principal cities of Europe, in India, South America, the ancient 
land of the Pharaohs, and the modern city in Australasia which 
takes its name from Lord Melbourne, Queen Victoria's first Prime 

And these exchanges were progressing so rapidly as to tax the 
resources of the proprietors, and with their growth arose new 
problems requiring the highest scientific treatment and introducing 
a new branch of engineering. 

P 2 



THE problems which faced the telephone engineers at the start of 
the exchange business were in the main solved by the light of 
the experience gained in telegraph engineering. Switchboards 
were adapted, the line construction was carried on from the older 
system to the newer. But as the subscribers grew in number 
and the transmission distances extended, the problems assumed 
an entirely new aspect. 

Intercommunication required that lines should be brought to 
a central office in order that they might be connected together, 
and the telephone engineer had to deal with hundreds or thousands 
where his predecessors had to deal only with units. So with the 
switchboard. The board which was adopted at New Haven per- 
mitted eight subscribers to be interconnected, the Williams and 
the other licensee. boards would work for a few hundreds, and the 
Standard switchboard gave a satisfactory service up to five or six 
hundred sometimes more. But the subscribers increased, and 
with the grov th in the number of lines the calls increased in greater 
proportion. The cost of operating was becoming a serious matter, 
but still more serious was the question of determining how to 
make the required connections at all. 

The two problems of outside or line construction and inside 
or switchboard work were attacked concurrently, but attention 
will first be given to the latter. 

The early switchboards, as described in Chapter XIV, had one 
point of access to a subscriber's line. The number of subscribers 
to which one operator could attend under the conditions then 
obtaining was usually limited to twenty-five, and rarely exceeded 
fifty, so that a connection between subscribers who were not both 
in any one operator's group would need the intervention of two 
operators, with some arrangement for making the connection 
between the two boards upon which were to be found the points 
of access to the two subscribers' lines. The Standard Board 



reduced this difficulty in part, since it was so designed that an 
operator could make direct connection with the board at either 
side, but beyond that the intervention of two operators was necessary 
together with transfer arrangements that became the more com- 
plicated as the numbers of subscribers increased. 

The solution was found by giving to each operator a means 
of access to every subscriber's line that is to say, by multiplying 
the points of access and the switchboard by which this was effected 
was called the ' multiple ' board. 

Next to the telephone itself the invention and development 
of the multiple switchboard must be regarded as the most potent 
factor in the telephone art. Without it, large exchanges would 
have been impossible, and the service through numerous small 
exchanges so slow and unsatisfactory that continued growth would 
have been unlikely. 

The germ of the multiple switchboard is found in the United 
States patent of Leroy B. Firman, No. 252,576, dated January 17, 
1882. The application was filed January 7, 1881, though the 
apparatus described was in operation nearly two years previously. 
Firman was the general manager of the American District Telegraph 
Company at Chicago. This company was introducing the Edison 
telephone in connection with their district system, and, finding 
much difficulty on the part of the operators in connecting subscribers 
through separate boards, Mr. Firman made some experimental 
designs towards the end of 1878, and about February or March 
1879 h e fitted up two boards at 118 La Salle Street, Chicago, 
upon each of which boards there were connecting points for all 
the subscribers to the exchange arranged in the manner illustrated 
in fig. 75, which reproduces fig. I of the patent. 

The prior state of the art and the scope of Firman's invention 
are clearly and concisely stated in his specification as follows : 

Prior to my invention the individual lines were grouped 
upon a single switchboard at the central office, or grouped upon 
two or more boards. In the latter case trunk lines were used 
when it was necessary to connect a line of one board with a line 
of another board. A large exchange was thus divided up into a 
number of exchanges, which could be worked together when occa- 
sion required, as one, by means of trunk lines between the boards. 
When the number of subscribers increased, so that a single switch- 
man could not do the amount of switching required, I gave the 
switchman an assistant. I soon found, however, that a single 
switchboard would not accommodate the number of attendants 
necessary to do the switching for an exchange of four or five hundred 

I find by the use of my new system of multiple switchboards, as 


hereinafter described, an exchange of a thousand or more subscribers 
may be successfully handled. 

My invention consists in providing two or more switchboards 
instead of one, as heretofore, and so connecting the several lines 
therewith that any two lines can be connected on either of the 
boards, and also apparatus, whereby attendants at a given board 
may without delay see what lines are connected at other boards 
than their own. 1 

Apparatus such as that referred to in the last few lines, whereby 
an operator could ascertain if a given subscriber were already 
connected, is an essential on any switchboard having multiple 

FIG. 75. Firman's (the first) Multiple Switchboard. 

connecting points. Firman's plan was of a very simple 
character. In the patent it is described as a ' dummy board or 
indicator ' upon which were recorded the subscribers' numbers. 
This board was placed in view of all the operators. The method 
of operating this primitive ' engaged test ' is thus described : 

Suppose lines a and c are connected at multiple board A, and 
lines b and d at multiple board A 2 , as shown by cords and plugs. 
The switchmen at the boards, immediately on making these con- 
nections, notify the attendant .at the dummy, who thereupon hangs 
up the shields or targets over the figures I and 3 and 2 and 4 ; and 
in the same manner, when any line is connected upon either of the 
multiple boards, the figure which indicates its number is covered, and 
a switchman, by glancing at the dummy, sees what lines are con- 

1 U.S. specification, No. 252,576, January 17, 1882. 


nected. For example, if the subscriber connected with plate 6 
were to ask for the subscriber connected with plate I, the atten- 
dant at board A', before making the connection, must glance 
at the dummy board, and in case he should see the target over 
figure i he would know that the line wanted was in use at another 
board, and instead of connecting plates 6 and i, he would notify 
the subscriber connected with plate 6 that the person wanted 
is busy. 1 

The next paragraph is of interest as illustrating the alternative 
methods of sending a disconnecting signal either communicating 
with operators by means of a circuit wire, as in the ' district ' system, 2 
or by the subscriber's direct line : 

The central office may be notified when the subscribers are 
through talking by the American district system, or by sending a 
current to line at either terminal station and tripping an annunciator 
number in the circuit at the central office. As soon as the signal 
to disconnect is received the switchman pulls out the plugs from the 
terminal plates or switches and immediately notifies the attendant 
at the dummy board to remove the targets. 1 

Mr. C. H. Wilson, who was indirectly associated with the District 
Telegraph Company, .says that the arrangement of the indicator 

did very fair work up to a certain point of the development of 
the business, but as the business grew and the lines became more 
numerous, . . . the necessity arose for a quicker method of deter- 
mining whether a line was in use or not, and the necessity of a 
larger number of switchboards was also apparent. 3 

The President of the District Company had some misgivings 
regarding the multiple principle, but Mr. Wilson reassured him on 
that point. He considered, however, 

that it would be necessary to develop some scheme by which an 
operator could determine instantly whether or not any given line 
was in use upon the other sections of the switchboard. I mean to 
say the sections other than the one at which the particular operator 
was at work. 3 

The necessity for the development of some such plan was also 
apparent to others, and Scribner was the first to devise a test 
system. This is shown in fig. 76 (fig. 6 of British patent, No. 4903 

1 U.S. specification, No. 252,576, January 17, 1882. 

2 Described in Chap, xxiii. 

3 Capital Telephone and Telegraph Co. Case, Circuit Court, U.S., Northern 
Dist., California, 1896, Supplemental Brief, p. 74. 


of 1879), where a local circuit, distinct from the line and normally 
open, is completed when a connecting plug is inserted in a ' block ' 
or spring-jack. ' The closing of the circuit by the insertion of a 
peg may operate indicators at each board in any well-known 
manner.' * The manner suggested in the patent was the electro- 
magnetic control of a valve in connection with a pneumatic tube 
where air under pressure should operate a piston to serve as an 
indicator. In this selection of a visual signal may be traced the 
line of development from Firman, whose ' dummy board ' or ' target 
board ' was looked at by the operator to obtain information as to 
the engagement or otherwise of a required line. This development 
is still more apparent in the United States specification, covering 
the same invention in greater detail. 

In fig. 10 sheet 3 [says the inventor] instead of a target for each 

FIG. 70. First Electrical Multiple Test System. (Scribner.) 

bolt, I have placed single targets, one for each subscriber's line, or 
series of bolts on an annunciator board in sight of the operators of 
all the boards. 2 

In other words, for the manually placed target of Firman was 
substituted a target electrically displayed when the operators made or 
broke a connection. While this alternative method was suggested 
in the United States specification, the plan contemplated for use 
was that the visual signal or target indicator should be brought on 
to the switchboard in front of the operators, multiplied in proportion 
to the number of lines and sections, and actuated automatically 
by the insertion or removal, as the case might be, of the plug with 
which the connections were made. 

The following description is from the British patent : 

1 British specification, No. 4903, 1879, p. 8, line-39. 

2 U.S. specification, No. 266,320, October 24, 1882 (application filed 
January 7, 1881). 


The insertion of a pin into either of the blocks of a series completes 
a local circuit, and so operates an indicator at each of the boards. 
In the figure a, b, and c are three blocks on as many different boards, 
forming a series belonging to one subscriber. Behind these blocks 
contact springs are placed in such positions that the insertion of a peg 
into either block deflects the spring into contact with a stud d. eisa. 
battery ; all the springs are connected with one pole of the battery, 
and all the studs d with the other pole. The closing of the circuit 
by the insertion of a peg may operate indicators at each board in 
any well-known manner. A very suitable way is to arrange an 
electro-magnet in the local circuit to operate on a valve and so admit 
air under pressure to a pipe leading to a small cylinder and piston 
indicators, of which one is provided in the vicinity of each block. 
When air is admitted to the pipe the pistons are raised to the tops 
of the cylinders, where they remain until, by another movement of 
the valve, the air is allowed to escape from the pipe, when the pistons 
fall by their own weight. 1 

The test of the pneumatic system was made in Chicago at the 
Western Union building, where compressors were in use for the 
pneumatic transmission of messages. 

In the course of an affidavit made in 1884 Mr. Scribner says that 
a portion of the invention was completed and described to others 
by diagram as early as February or March 1879, and that a test of 
the pneumatic duplicate switchboard system took place in the 
summer of 1879. 

Mr. Wilson, in conjunction with Mr. Clark C. Haskins, developed 
another plan, shown in the United States patent, No. 266,287, dated 
October 24, 1882, the application for which was filed October 23, 
1879. After an introduction reciting the prior usage which has 
already been explained, the patentees say : 

The object of the present invention is to provide a means whereby 
several duplicate switchboards may be employed, to each one of 
which all the telephone stations are connected, and so arranged that 
when any one line is in use or occupied at one of the switchboards, 
that fact will be rendered apparent at all the other switchboards, 
thus preventing any confusion between the different operators at 
the several switchboards, and also preventing any accidental 
crossing of the lines. 2 

Fig. 77 follows closely the sketch of Firman (fig. 75), but instead 
of having an indicator board upon which targets were placed by 
an assistant when a line was engaged, provision is made for the 

1 British specification, No. 4903, 1879, p. 8, line 32. 

1 U.S. specification, No. 266,287, October 24, 1882 (application filed 
October 23, 1879). 


indication of an engaged line by an electrical test. The explanations 
of the drawings given in the specification are as follows : 

Figure [77] is a view of several duplicate switchboards, there being 
three in this instance, arranged according to our invention. Fig. [78] 


is a perspective view of an improved switch such as we prefer to 
employ in working our invention. Fig. [79] is a sectional view on the 
line xx of Fig. [78]. Fig. [80] is a sectional view on the line^y of 
fig. [79]. [K'] Fig. [78] is a perspective of one of the plugs removed 
from the switch. Fig. [81] is a diagram of the. call circuit employed 
in conjunction with the apparatus at the central office. 1 

It will be noted that the switch contains the essential features of, 
and performs the same operations as, the modern spring-jack. No 
insulating material being used in its construction, it is necessarily 
mounted on the board in three parts, d', M', and d. M' consists 
of two separate parts, one fixed (M'), the other (M) sliding 
within a suitable guide piece. M has a projecting bar with collars 
at either end and a helical spring (m) over it. When no plug is 
inserted the spring forces the sliding piece into contact with the 
stud d. Holes are bored through M and M' adapted to take a plug, 
but when the slide is in contact with the stud these holes do not 
exactly register, but the movement of the slide is so limited that 
enough of the hole remains uncovered to receive the point of the 
plug, and upon its further insertion the plug, acting as a wedge, 
overcomes the operation of the spring and forces the slide M out of 
contact with d. d' is called a ' try piece.' The combination is 
therefore comparable with a single circuit multiple jack, M' repre- 
senting the spring, d the contact, and d' the frame or testing point. 

The method of wiring is clearly shown by fig. 77, where ' J is a 
line wire from a distant telephone station entering the central 
office. This wire is connected first to the metal of the switch D 
at the board H. This switch being closed, the circuit is through 
it to the contact plate d, and from this to the switch D on the 
board H'.' The operation is repeated on the board H 2 , where the 
wire continues to ' trying post ' d' on that board and to each ' trying 
post ' on the preceding boards. 

The circuit may thus be described in modern terms as com- 
mencing with the spring of a multiple jack through the contact on 
that jack, and similarly through successive jacks to the last section, 
where it is connected to the frame of that jack and successively to the 
frames of each jack of the series to the first section. 

Connection between two jacks is made by two metallic plugs, 
K and K', which are connected by a flexible cord, and the insertion 
of the plug in either jack breaking the contact as already described. 
It being assumed that A and D lines are connected together, 

if an operator at either of the other switchboards has a call for either 
of the lines represented by the switches A and D, he will first proceed 
to ascertain if those lines are in use at either of the other switch- 
boards, and this he will do as follows : 

1 U.S. specification, No. 266,287, October 24, 1882. 


The central office is provided with a magnetic call bell and circuit 
(shown in fig. [81]). The call bell may be one upon a telephone, for 
convenience. We will suppose that the operator on the switchboard 
H' has received a call for the subscriber connected with the lines of 
the series of switches D. He tests the question as to whether this 
line is in use upon either of the other switchboards by inserting a 
plug from the circuit at fig. [81] into the trying plated' on his switch- 
board H'. If he gets circuit he knows that the line is unoccupied, 
as in this case he would be in circuit shown at fig. [77] ; but if the 
switches A and D on the board H were connected, as previously 
supposed, he would not, by trying at the board H', get the circuit, 
because the connection would be broken between the switch D 
and the contact plate d on the board H' ; nor would he get circuit 
if the switch D on the board H 2 were in use, because in that case the 
contact plate d and switch D on that board would be disconnected, 
and there would be no circuit from the trying plate d' on board H' 
beyond the contact plate d on H 2 . Thus it will be seen that the 
circuit is always cut out behind the operator at each switchboard 
when any switch is in operation. 1 

It will be understood that in ' getting circuit ' a signal is made. 

In the preceding description the operation of making a connection 
has referred to only one hole of the switch, but, in the words of the 

it will be observed that we provide each switch with two holes. 
While this is not absolutely necessary, we find it convenient, for 
the reason that it may be desirable to connect two stations with the 
telephone at the central office before connecting them with each 
other, and two holes being provided, the plugs connecting the 
switches together may be inserted before the plugs belonging 
to the office telephone are removed. 1 

Another point of detail to which attention may be drawn is the 
provision made for retaining the plug when inserted in a switch. 
On reference to K', fig. 78, it will be seen that a slot or notch is cut 
in the plug. The patentees say : 

A plug when inserted should be turned so that the slide piece or 
lever M may fit into the notch of the plug. The plug is thus held 
securely in place. 1 

It has been stated previously that Firman was manager 
of the district telegraph system at Chicago, which concern was 
connected with the Western Union Company, and that the 
Western Electric Manufacturing Company was associated with the 
Western Union Company. Their factory was at Chicago, and 
various new devices which were required as the exchange progressed 

1 U.S. specification, No. 266,287, October 24, 1882. 


came under their notice. Their technical staff included C. E. Scribner, 
who joined the Gold and Stock Telegraph Company at Chicago 
in November 1876, doing then occasional work for the Western 
Electric Manufacturing Company, and in the summer of 1877 was 
transferred to the latter company. To Mr. Scribner Mr. Firman 
described his system, and the apparatus was constructed by the 
Manufacturing Company. In the course of evidence in a law 
case, Scribner relates that he 

personally had to do with this work, saw the apparatus constructed 
and installed, and witnessed its successful operation. 1 

From the same source there may be taken with advantage the 
evidence of this eye-witness as to the position of the exchange 
system in these early days and the problems with which the pioneers 
were faced. 

The value of the telephone exchange came to be appreciated 
by the public very soon, and the growth of the telephone exchange 
gave to its promoters surprise following surprise. A switchboard 
designed and well calculated to provide for the estimated growth 
of years would be found of too small capacity for the needs of the 
time when completed the growth of the business between the date 
of the order and that of completion being sufficiently great for this 
result. The single operator of the early exchange was first given an 
assistant, other operators and assistants were added, and a difficult 
problem soon presented itself to the pioneers of telephony. With 
the increasing business and the necessary employment of a number 
of operators, it was found that each operator must be assigned a 
limited number of calls. 2 

After describing the single form of switchboard, Mr. Scribner 
continues : 

With the earlier growth of the telephone business, and up to the 
time when three or four hundred subscribers were connected with 
the system, this plan of operation succeeded fairly well. It was 
apparent to those studying the question and directly dealing with 
the problem that with increasing business the limit of the capacity 
of this system would soon be reached. The problem presented to 
those beginners in the telephone business was to provide for the 
interconnection of any two lines of an extremely large number of 
lines ; to avoid the possible connection of three lines together at 
any one time ; to provide for the prompt disconnection of any 
two connected lines, and to give to each operator of the system the 
apparatus with which to accomplish this result. That is to say, each 

1 Capital Telephone Case, Circuit Court of United States, Northern District 
of California, 1896, Complainants Record, p. 20. 

2 Ibid. p. 19. 


operator of the exchange must be provided with facilities for con- 
necting any one of the lines assigned to her with any other line in 
the entire exchange. But she must be protected against making 
a connection with any line which was already in service, whether 
such line be one assigned to her or to any other operator of the 
exchange. The problem of interconnecting lines for communication 
originated with the advent of the telephone, and was an entirely 
new one to electricians and engineers. Many different solutions 
of the problem were offered and many tried. Failure followed 
failure, and in the first months of 1879 perfect chaos existed in the 
larger telephone exchanges of that time. No single switchboard 
could be constructed which would be large enough to provide for 
the switching of the exchanges at that time ; increasing the number 
of operators increased the difficulty, for the reason that it resulted 
in a division of the lines of the exchange into smaller groups and 
increased the number of groups, thus increasing the difficulty of 

FIG. 82. Multiple Switchboard double ' block ' method of connection (1879). 

establishing a connection between a member of one group and a 
member of any other group. 1 

In 1879 Scribner was in London for the purpose of taking out 
the patent already referred to, which included several features 
in connection with telephones which his company had improved. It 
was partly a communication from George D. Clark, Milo G. Kellogg, 
and George B. Scott. It included an improvement on the Edison 
transmitter, and the circuits of the Edison sub-stations as they 
were introduced by the London Edison Company. But the principal 
feature of this patent was the multiple switchboard. 

It will be seen that the circuit of fig. 82 2 has a resemblance to 
the plan of Wilson and Haskins, but instead of passing through 
one switch on its way out and returning to a stud of that 
switch, two switches are used. The switches are of the ' jack- 
knife ' form a blade pivoted in a frame and resting on an insul.itrd 

The purpose of the arrangement shown in fig. 82 is thus 
described : 

When the number of subscribers is large it becomes necessary to 

1 Capital Telephone Case, Circuit Court of United States, Northern District 
of California, 1896, Complainants Record, p. 19. 

2 U.S. equivalent, No. 321,390, June 30, 1885. 


divide them into sets, so that one attendant may attend to the calls 
of one set, another to those of another set, and so on. I then have 
a separate board for each set, and on the board are the call indicators 
of the set, and those only ; but the line wire of each subscriber is 
brought to the board, for a subscriber belonging to the set may 
desire to speak with another subscriber out of the set, and it is the 
duty then of his attendant to make the necessary connections. But 
a difficulty arises when the subscribers are thus divided into sets 
having different attendants, for when a subscriber is called for at 
one board the attendant does not know whether he may not be 
already engaged and coupled for conversation at another board. 
To avoid inconvenience arising from this cause I provide for each 
subscriber two blocks on each board, an upper and an under one ; 
they are coupled in series, the upper block on the first board makes 
contact for the upper block on the second board, and this for the 
upper block on the third board, and so on. Then, at the end of the 
set of boards the connections are continued through the lower 
blocks, each in succession until last in the series, the lower block 
upon the subscriber's own board is arrived at where (as previously 
described) the connection is made with the call indicator and so to 
earth. Now, for the purpose of calling a subscriber the peg is 
inserted into the hole in the lower block, and so will not interfere 
with any connection which may have been previously made at 
either of the upper blocks ; but when such a connection exists the 
call will not pass, the connection with the subscriber's line having 
been previously broken. 

It is obvious that for upper and lower, right arid left may be 
substituted. The arrangement of the connections above described 
to be used when multiple boards are used is clearly shown by fig. [82] 
of the drawings. 1 

The connections are, in fact, so clearly shown as to make it 
unnecessary to quote the description in further detail. 

The design shown in fig. 83 is stated by Mr. E. M. Barton in a 
letter 2 to Mr. F. R. Welles, dated August 2, 1883, to be Kellogg's 
contribution to the patent. The description of this system is as 
follows : 

Or in place of this [the fig. 82] arrangement another requiring 
but one block to each subscriber on each board may be adopted. 
Thus, a small finger key or contact maker may be provided in 
connection with each subscriber's block, and the attendant when 
he desires to know if a subscriber is at liberty places his finger on 
this key or contact maker ; if the subscriber be occupied the contact 

1 British specification, No. 4903 of 1879. 

2 The occasion of this letter was the then recent introduction of a multiple 
test system on a trunk exchange at Coleman Street, London. This was 
independently designed by Mr. F. B. O. Hawes and was subsequently applied 
to a subscribers' board at the Chancery Lane Exchange (Journal of the 
Institution of Electrical Engineers, xxv. 367). 


thus made completes a circuit in which is a battery and a small bell 
or ' buzzer ' ; and the sounding of this bell or buzzer indicates that 
another connection ought not to be made. This circuit, however, 
only becomes complete when one of the knife-like blades, corre- 
sponding to this line wire, is pegged out on one or other of the boards, 
the blade being by this movement brought against a contact point 
and so completing the circuit ; but when by the removal of the 
peg the blade is permitted to spring back this local test circuit is 
broken, and although the finger key or contact maker may be pressed 
down the ' buzzer ' will not sound. 1 

The arrangements to be made when this system of local circuits 
is applied to ascertain when a subscriber is at liberty is illustrated 
by fig. [83]. Here portions of three boards are represented, and 




r . 

y~o cr-) 

LEt ;_ 






5* cr-o -dj 


FIG. 83. Kellogg's Multiple Test. 

two blocks, belonging to different subscribers, are indicated on each 
board. The blocks employed are similar to those shown at figs. [85] 
and [89], except that in each case an insulated peg is provided upon 
the blade which, when the blade is lifted off its contact screw by 
the insertion of a peg, raises a spring a into contact with a metal 
stud b. The three studs b on the different boards appertaining to 
each subscriber are connected by an insulated wire c ; two of these 
wires are seen in the drawing. 

The springs a of all the blocks on all the boards are connected 
to one common wire d. Over each block there is a finger key e, 
which when pressed down makes contact with the corresponding 
stud b. Each board has its battery /, and one pole of the battery 
is connected with all the finger keys e on the board, whilst the other 
pole of the battery is connected through the electric bell or ' buzzer ' 
g (of which also there is one to each board) to the common wire d. 
The attendant at either board, when wishing to ascertain if a 
subscriber is disengaged, presses down the finger key corresponding 
to this subscriber's block, and thereby causes the bell or ' buzzer ' g' 
belonging to the board to sound if at either of the boards the 
subscriber's wire is already coupled, but otherwise the bell or 
'buzzer ' does not respond. 1 

1 British specification, No. 4903 of 1879. 


The test system of Haskins and Wilson l is a circuit starting 
from earth at the exchange through a battery, a bell, a 
flexible cord ending in a plug inserted in the ' try piece ' d ' , the line 
(if disengaged) and earth at the subscriber's office. The circuit 
is thus complete and the bell sounds, indicating a disengaged, line. 
In fig. 83 (Kellogg's) the bell circuit is local and distinct from the 
line. There is a ' buzzer ' (the equivalent of Haskins and Wilson's 
bell) at each section, and the ' buzzer ' sounds as an indication 
of an engaged line. The test circuit is completed by a finger key 
attached to each jack-knife switch. In the letter from Mr. E. M. 
Barton previously referred to it is said : 

Scribner states that Mr. Kellogg's first plan for this style of 
switchboard was a separate testing plug at end of a cord [similar to 
Haskins and Wilson] , and that when he was describing the apparatus 
to Carpmael for him to lodge the specification he described to him 
this identical arrangement ; but it would seem that Mr. Carpmael 
did not think it worth while to put in a separate description of it in 
the patent, and Scribner thinks that the use of the term ' contact 
maker ' in the English specification was employed by Carpmael to 
make the description include such a ' contact maker ' as the special 
cord and plug. 

It seems probable, however, that Mr. Carpmael did more than 
cover two specific plans, that he adopted a phrase which should 
include any effective test by making contact and thus closing a 
circuit. If so it was an example of an important addition by a patent 
agent who had perceived the scope of the invention, for neither 
the finger key nor the special cord and plug survived as the 
means for obtaining information by the operator as to the engaged 
or disengaged condition of the line. This British patent was 
not contested throughout its life, and the words added by Mr. 
Carpmael were an important aid to that result. The completeness 
of the invention is remarkable when the early date of the industry 
is considered. In London, for example, a telephone exchange had 
been established only about three months when the specification 
was deposited at the Patent Office. 

The modifications needed for metallic circuit lines are provided 
for in fig. 84, where are shown the changes in the ' knife-like blocks ' 
which have to be made when ' it is advisable to arrange telephonic 
circuits with two line wires in place of completing the circuit by 
means of earth connections.' 2 

The jack-knife switch shows a considerable advance in design 
over the switch illustrated in Haskins and Wilson's specification. 

1 Figs. 78-81. 

2 The equivalent U.S. specification is Scribner's 266,319, October 24, 1882. 



But it would appear that their design was not put into use, for in 
their paper read before the American Electrical Society in December 
1879 they say : 

A connecting plate devised by Mr. C. E. Scribner of the Western 
Electric Manufacturing Company of Chicago, which has proved 

/.I'M Lin.* 

FIG. 84. Metallic Circuit Multiple (1879 patent). 

of practical utility in this connection, has b?en adopted by the 
several companies using the multiple system. It is known as the 
' jack knife.' 1 

Insulation in Haskins and Wilson's design was obtained by 
separation, and the three parts were individually mounted on the 

board. The jack-knife switch is 
compounded of metallic and in- 
sulating parts, and forms a struc- 
ture fixed to the board by one bolt, 
which is also the connecting point 
for the line. The purpose served 
is the same, the line coming to the 
block (figs. 85 and 86) with which 
the blade a' is metallically con- 
nected and leaving by the screw 
a" 'which, though mechanically a part of the structure, is insulated 
by an ebonite sleeve. 

The jack-knife switch is more fully described in Scribner's U.S. 
patent, No. 293,198 as follows : 

Figure [87] the original form of a part of my device is a flat 
piece of metal, N, slit as shown, and provided with holes x and y 
and contact points C and D, the latter of which, D, is insulated, as 
shown, by hard-rubber bushing. The lever B, (shown by Fig. [88] 

1 Journal of the American Electrical Society, 1880, p. 48. 

FIG. 86. 

FIG. 85. 

Jack-knife Switch. 



in detail,) in combination with spring I and frame or back E, and 
pivoted to E, as shown at r, Fig [89], takes the place of the single 
piece N. In the single piece N advantage is taken of the elasticity 
of the metal. Frame E is provided with two holes, x and y, and 

FIG. 88. 


T^ x^"^ 

FIG. 87. FIG. 89. 

Jack-knife Switch in Detail. 

a slot, in which lever B rests. The inner edge of the lever B projects 
so as to come within the edges of the plug holes x and y, as shown 
by dotted line, and when a plug is inserted in either hole the lever 
is forced down, as shown by lower dotted line, and the points of 
contact D C are separated, C taking the position shown as indicated 


FIG. 90. Jack-knife Switch (section and connections). 

by C'. a is a metallic plate screwed to hard-rubber block b, and 
thereby insulated from the other portions of the switch. 1 

Giving evidence in a patent suit, Mr. Wilson said that the exact 
time when the invention of Haskins and himself was made could 

1 U.S. specification, No. 293,198, February 5, 1884 (application filed 
August 23, 1879). 

Q 2 


not be stated, but it was some weeks, and perhaps a few months, 
prior to the making of the application for the patent. 1 

I remember distinctly [added Mr. Wilson] that we were in no 
hurry to apply for the patent until we became suspicious that 
other parties were about to apply for a patent upon what we 
regarded as our invention. 2 

The reference indicates the active efforts which were being made 
by various inventors to effect improvements. Scribner, Kellogg, 
Haskins and Wilson were all in close contact with Firman, and any 
rivalry which may have existed must have been entirely of a personal 
character since ah 1 their patents were .assigned to the same firm. 3 
But any rivalry of this kind only serves to indicate that these 
inventors were fully conscious of the importance of a test system. 
Mr. Kellogg, in the course of a conversation which I had with him 
in 1891, said that all produced their methods within a fortnight in 
October 1879. But whilst this may refer to Scribner's fig. 4 of 
1879 (fig. 82), to Haskins and Wilson's and to Kellogg's own 
method, the evidence is clear that Scribner's fig. 6 (fig. 76) ante- 
dated that period by several months, and thus stands as the first 
multiple testing system. But it was not used in any working 
exchange. In Chicago at this time they were working both on the 
American District system and on the direct line system. The 
Haskins and Wilson method was adopted for the open lines in the 
American District Exchange, and Scribner's fig. 82 went into service 
in a number of the exchanges with direct lines. Kellogg's plan 
(fig. 83) would appear to be a modification of the separate test 
circuit of Scribner's fig. 6 (fig. 76), this separate circuit however 
remaining open until closed by the testing operation of depressing 
a finger key or contact. Kellogg's method, like Scribner's pneumatic, 
was not adopted in practice. Nor need this occasion surprise. 
It would naturally be preferred to economise the wiring and avoid 
complication by using the line circuit so far as possible as in Scribner's 
fig. 4 (British) (fig. 82) and Haskins and Wilson's ' try circuit.' 

Du Moncel, who was the first to give a popular and scientific 
account of the invention of the telephone, was the editor of La 
Lumiere lectrique, and, so far as Europe is concerned, it is to this 
publication that we must refer for contemporary descriptions of 
the multiple switchboard. These appeared in 1880, and abstracts 
of them are included in the fourth (French) edition of Du Moncel's 
Le Telephone, published in 1882. It is to be observed that no 

1 October 23, 1879. 

8 Capital Telephone Case, Circuit Court of United States, "Northern District 
of California, 1895, Complainants Record, p. 73. 

3 The Western Electric Manufacturing Company. 


attempt was made to translate such technical terms as ' try circuit,' 
' try plate,' or ' jack knife ' which last, however, appeared as 
' jack knif.' In the subsequent English edition of Du Moncel's 
work no reference is made to the switchboard apparatus. 

The articles in La Lumiere Electrique cover much the same ground 
as that in the Journal of the American Electrical Society (1880), 
which was prepared by Messrs. Haskins and Wilson at the request 
of the Publication Committee of that Society in order to complete 
the brief description given at the annual meeting held on 
December 10, 1879. This description was probably the first public 
explanation of the multiple switchboard. The record thereof in 
the society's journal was evidently not written by a telephone 
expert, but the final remarks indicate that the general idea was 
accurately absorbed : 

Mr. Charles H. Wilson gave the members a description of a 
new switchboard of his invention, to be used as a governor for wires, 
whether telephonic or telegraphic. He drew diagrams on a black- 
board illustrating the process. One principle was the introduction 
of a polarized relay, to operate when a subscriber caUed, but would 
not sound when the operator responded, and a plug was so arranged 
that the operator had control of all the tie bolts at once. The 
board seems to be in ever}' way practicable, and meets valuably 
a demand that has long bothered operators. The arrangement 
is such that an operator can readily tell when another is using the 
desired wire. 1 

The various types of switchboard described in Chapter XIV 
were gradually discontinued, except the standard switchboard, 
which was retained for small offices. For large offices the multiple 
board, the development and principles of which have been described 
above, was becoming general but not yet universal. 

The terms ' large ' and ' small ' being relative, it becomes 
necessary to define them, and this may best be done by a quotation 
from a printed document of the period. In 1883 the manufacturers 
issued a ' Descriptive Circular of the Multiple Switchboard for 
Telephone Exchanges.' 

The opening paragraphs of this circular are : 

The two most important requisites in switchboards for large 
telephone exchanges are speed and economy. 

It so happens that these two requisites go together ; for whatever 
tends to reduce the amount of work to be done also enables the 
operators to do it quickly, and therefore diminishes the cost of the 

1 Journal of the American Electrical Society, 1880 (Appendix). 


The requisites of speed and economy are met with in a high 
degree in the multiple switchboard. 

In this circular it is stated that : 

This system is now in use, or soon will be, in the following cities : 
Minneapolis with 800 subscribers ; Milwaukee with 900 subscribers ; 
Nashville with 800 subscribers ; Kansas City with 500 subscribers ; 
Melbourne, Australia, with 700 subscribers ; London with 500 
subscribers ; Baltimore with 1600 subscribers ; Washington with 
1000 subscribers ; New Orleans with 1500 subscribers ; Indiana- 
polis with 1000 subscribers ; Liverpool with 1000 subscribers ; 
Columbus with 600 subscribers ; Toledo with 500 subscribers ; 
Dayton with 800 subscribers ; Memphis with 600 subscribers ; 
Peoria with 500 subscribers. 

From this it may be gathered that an office became ' large ' 
when it approached 1000 subscribers, and that it was then becoming 
generally recognised to be sound practice to instal the multiple 
board in the event of new switchboards being required for a 
central office that was expected to exceed 1000 subscribers. 

The principle upon which the multiple board is based has been 
fully described, so that it is unnecessary to follow the circular in 
detail, but the method of operation and some other points are so 
clearly and tersely related in it that some quotation is permissible. 

An operator has a certain number of subscribers whose calls 
she answers. When one of these subscribers has called for a 
connection with any other subscriber of the exchange, the operator 
who takes the order makes the connection of the two lines instantly 
and rings the bell of the subscriber wanted. 

When the conversation is finished, the same operator who con- 
nected them together receives the notice by the subscribers to 
disconnect, and then disconnects the lines. This is all done by 
the operator without moving more than one step. 

It is hard to imagine a system more simple or more easily 
manipulated. It is in consequence of its simplicity that it is so 

As many switchboards are required in an exchange as there 
are times two hundred subscribers. For instance, with one thousand 
subscribers, five switchboards would be necessary. 

Several different arrangements of the spring-jacks and annuncia- 
tors, with reference to their disposition on the board, have been tried. 

The annunciators have been placed all above the spring- jacks, 
all at the two sides, and part above and part at the two sides. 

None of these arrangements, however, seems to possess the 
advantages which are secured when the annunciators are placed 
across the boards below the spring-jacks, with the operator's key- 



board below the annunciators, and the shelf for holding the cords 
and pulleys above the spring-jacks. 

One important advantage of this general arrangement is that it 
allows the placing of the boards together, end to end, thus enabling 
an operator at any board to make connections between the spring- 
jacks upon her own board and those upon the adjacent board, in 
cases where those spring-jacks are nearer than the correspond- 
ing ones upon her own board. . . . 

The accompanying cut [fig. 91] shows the jack as used on the mul- 
tiple board. The frame 
of the jack is inserted at 
the back of the board and 
screwed fast by a lug 
which forms a portion of 
the frame. 

The phosphor-bronze 
spring, and the contact 
point upon which it FlG - 91- Multiple Jack (1883). 

normally rests, are insu- 
lated from the frame of the jack by hard rubber bushings and 

The connections of the jacks are all made at the back of the 
board, and, to facilitate this work, phosphor bronze connecting 
pieces are extended from both the insulated contact point and 
spring, bringing all the connecting points together in one plane 

When a plug is inserted into the jack [fig. 92] the phosphor-bronze 

FIG. 92. Multiple Jack and Plug (1883). 

spring is lifted from the insulated contact point, breaking a con- 
nection therefrom, and a new connection is established to the tip 
of the inserted plug upon which the new spring now rests, and 
thence to the cord connecting with the plug. 

A connection, or cross, is also established between the phosphor- 
bronze spring and the frame of the jack, through the medium of the 
plug. It is this cross as thus established (and which is more fully 
described hereafter) which provides for the operators' test. 

The accompanying diagram of circuits [fig. 93] shows three spring- 
jacks of a line placed respectively on three multiple switchboards. 

The circuit through these three jacks and the annunciator is 
from the subscriber's line through the lightning arrester to the 
spring of the first jack ; from the contact point upon which this 


spring rests to the spring of the second jack ; from the contact 
point upon which the second spring rests to the" spring of the third 
jack ; and from the contact point upon which this last spring rests 
the circuit is through the annunciator belonging to that line, and 
thence to the ground. 

As is evident, the annunciator may be placed at any of the three 
boards, but at whichever board it is placed, its magnet must be 
included in the circuit between the last spring-jack and the ground. 

The springs and contact points included in this circuit being, 
as above described, insulated from the frames of the jacks by hard 
rubber, no connection exists between the line circuit and the frame 
of any jack during its normal condition that is when not in use. 

The frames of the three jacks are all connected together, but 
normally to nothing else. 

FIG. 93. Line Circuit and Test Wire of Multiple Board (1883). 

The line circuit through the system, as shown in the diagram, 
is lettered A. The wire which serves to connect together the frames 
of the jacks is lettered B. ./ 

The insertion of a plug at any of the three jacks, as shown at 
the second jack, lifts the phosphor-bronze spring free from the 
contact point upon which it normally rests, thus disconnecting 
the annunciator and ground, and establishing connection between 
the line and the cord connecting with the plug, as well as the 
cross-connection between the line and the frame of the jack. 

The frames of all the jacks of the same number being connected 
together are consequently all crossed with the line. 

An operator at any other board may therefore determine the 
fact of a cross between the jack frames and the line by touching 
the frame of the jack at the front of the board with the plug which 
she is about to insert into the jack to complete a connection. 

The keyboard connections are shown in fig. 95. The first 
impression of those familiar only with more modern boards is 



likely to be that the block has been inverted in the course of prepara 
tion for the press. The reader, however, may be reassured. In 
the early multiple switchboard the plugs and cords were not upon 
the keyboard, but were placed above the jacks and indicators, as 
will be observed more readily from subsequent illustrations. 

The circuit and operation is too familiar to need quotation 
from the description, but an illustration of the earlier cabling 
methods will be of interest. 

Heretofore considerable trouble has been experienced in cabling 
the connections from the multiple switchboard, occasioned by 
the wires accumulating at the sides of the board in large cables, 
covering over the spring-jacks and rendering them inaccessible. 

Test Battery 

FIG. 95. Keyboard Connections Multiple Switchboard (1883). 

To obviate this difficulty spring- jacks are now arranged in 
divisions of twenty, leaving space for the cable to be run perpendi- 
cularly between these divisions. Perpendicular strips dividing 
the boards are placed in these divisions upon which to run these 

The boards are perforated, and the wires in cables, of no larger 
than twenty wires each, pass through the perforations and are 
supported thereby. 

The wires by this method are evenly distributed over the entire 
surface of the board, the cables coming out from the board at 
regular intervals, and at no point is there an accumulation of 
greater than twenty wires in one cable. 

All the spring- jacks of the system are accessible and can be 
removed in case it becomes necessary to replace them. 

The accompanying diagram [fig. 96] shows a cross-section of the 
multiple board, the dividing strip or cable support, and the cables 
passing through it, as described. 

The circular has an illustration of ' The Multiple Switchboard 


for Indianapolis ' reproduced in fig. 97. It will be understood that 
the legs and the overhead cord structure are omitted. 

Of some of the other boards mentioned in the circular as being, 
or about to be, in use, illustrations when fitted in the various 
exchanges are reproduced in figs. 94, 98, and 99 from the sources 
mentioned ; and from the London Electrical Review of October 18, 
I884, 1 an illustration (fig. 100) of the multiple switchboard in- 
stalled at Liverpool, the first in active operation in England. 
The London board referred to in the descriptive circular was 

FIG. 96. Cabling System Multiple Switchboard (1883). 

supplied to the London and Globe Telephone Company, whose 
service was restricted by the action brought against them by 
the United Telephone Company for infringement of the Edison 
patent by the use of the Runnings transmitter. 

The reason why the plugs were placed above the boards was 
that in the then normal height of the keyboard from the floor 
there was not sufficient room to take up the slack of the cord of 
the length required to cover the jack field. From a new issue of 
the ' Descriptive Circular ' in the following year (1884) it is to be 
inferred that there were very certain drawbacks to the overhead 
system. The 1884 board had other important changes, but we 
will first quote the paragraph relating to the cords. 

1 Vol. xv. p. 312. 





^tl: :!::! ^:ti: isi : 1 1; '<1 


FIG. 98. Multiple Switchboard, Melbourne, Australia. (From the 
Town and Country Journal, September 10, 1887.) 


FIG. 99. Multiple Switchboard, Melbourne, Australia, in Operation. 
(From the Town and Country Journal, September 10, 1887.) 



The cords are all below, thereby avoiding the use of several 
pulleys and the consequent squeaking and rattling noise caused by 
them. Having the cords separated also gives more room for the 
weights under the table, and less chance for trouble in moving 
them. The weights are about one-third as heavy as on the over- 
head cords, therefore the cords will last longer. 

The illustration fig. 101 forms the frontispiece of the 1884 

FIG. 101. Multiple Switchboard (1884). 

'Descriptive Circular/ and indicates important changes in details 
apart from the restoration of the plugs to the keyboards, for, as 
it will be noted, there are two shelves. The board was consequently 
known as a ' double-decker.' Originally the only jacks upon the 
board were those in the multiple field. These were arranged in blocks 
of one hundred. As the sections were designed for 200 subscribers, 
it follows that the operators answering calls at one section were 
in competition for the jacks in a space of two blocks. This is 


a drawback which would be very obvious in practical work, and 
it was remedied by the introduction of ' answering jacks.' The 
' multiple ' multiplied jacks on each section. The answering jack 
was an extension or refinement of the same principle, inasmuch as 
the jacks relating to the subscribers at a section were duplicated 
on that section, one appearing in the multiple field above, the other 
in the answering field below. Interference between operators in 
the multiple field was thus reduced to the selection of the jack 
for the called subscriber ; that of the calling subscriber was directly 
before the operator to whom it was allotted. 

The pairs of plugs were divided, one, the answering plug, being 

FIG. 102. Multiple Spring-jacks (1884). 

placed on the lower shelf, and its mate, the connecting plug, on 
the upper shelf. 

There were important changes in the component parts of the 
board, which the circular shall describe : 

The new spring-Jacks are mounted upon narrow strips of hard 
rubber, so that very much better insulation is secured between the 
parts (see cut of spring-jacks) [fig. 102]. They are very much 
smaller in size, occupying only about one-half the space on the 
face of the board formerly occupied by the larger size. Five of 
these strips constitute a section of one hundred jacks. 

The annunciators are also much condensed, and are furnished 
in sections of ten on a strip one inch wide by fifteen inches 

Exchanges will grow, and the boards are made large enough to 
admit of any growth that is liable to occur. 



The use of ' half sections ' or dummies,' as they were some- 
times called, probably preceded 1884, but they are first mentioned 
in this circular. After describing how the jacks at the beginning 
of No. 2 board are available for the use of the operator at the end 
of No. i board and so on, the circular says : 

This principle may be carried to the extent of placing a 
half section of board at each end of the system of multiple 
boards, thus giving the operators at the beginning of the first 
board and at the end of the last board the same advantage in this 

The following exchanges with the number of subscribers against 
each are quoted in the 1884 circular as having adopted the multiple 
board in addition to those previously quoted from the circular 
of 1883 : 

New York 

Buda-Pesth, Hungary 

Gothenburg, Sweden 




Lawrence, Mass. 



2 exchanges with 2,000 subscribers. 




The illustration fig. 103 represents the new type of board 
(described in the 1884 circular) as fitted in the Boston exchange, 
and is reproduced from the New York Electrical Review of 
September 26, 1885. 



AT the time that the telephone was commercially introduced, 
telegraph engineering had reached a high stage of development. 
For land lines galvanised iron wire was almost universally used. 
The wires were strung overhead and attached to poles or fixtures 
by means of insulators of types which varied with localities : those 
of glass being very generally used in the dry climates of America, 
whilst porcelain ones were almost universal in Europe. Submarine 
cables were numerous, and the scientific principles governing their 
operation in the transmission of telegraphic signals were subjects of 
constant investigation. The conductors in these cables were of 
copper, and the insulation was gutta-percha or india-rubber. Land 
cables when used followed the submarine type of construction, but 
for underground conductors in congested districts gutta-percha 
covered wires were drawn into iron pipes in groups rather than made 
into cables. 

The line construction methods of the telegraph system were 
naturally adopted by the early telephone engineers ; but, as the 
installations grew in number and the transmission distances extended, 
new problems were constantly being met with which required for 
their solution important modifications in established practice. 

For telegraph work, experience was available which permitted 
estimates to be made with reasonable accuracy so that telegraph 
lines could be planned and built with due regard to requirements 
and results. But there was no experience of telephone exchanges ; 
and, if the promoters had estimated for the numbers which actually 
came, they would certainly have been accused of rashness and, as 
certainly, they would not have obtained the capital to carry out 
their plans. In consequence there was, in the early days, an element 
of haphazard in design as well as an effort at cheapness in construc- 
tion which, whilst the subjects of criticism by telegraph engineers 
of the period, were really inevitable and, in the long run, not disad- 
vantageous, because the quicker and cheaper construction permitted 



earlier use and more prompt and widespread demonstration of the 
advantages of the service, whilst there was less compunction about 
sacrificing such lines and substituting others when experience showed 
that it was desirable to do so. 

The route of a telegraph line and the office to which it should 
be led were to some extent under the control of the authority 
constructing it, and the office might be chosen with special regard 
to getting the line or lines into it. But the essence of the telephone 
exchange business was that the new means of communication should 
be available at the place of business or residence of any member of 
the public, so that the lines must start from points which could not 
be chosen by the engineer but must, nevertheless, be provided for. 
These points were naturally more numerous in congested business 
and residence areas, so that the telephone engineer on the line side 
of his work was faced with just the same problem as on the exchange 
side how to provide facilities for the growth in numbers which 
rapidly came. 

While one terminal of the line the user's office was beyond 
the engineer's control, the other terminal the exchange was his 
own choice ; and, so soon as a little experience had been gained, this 
was usually placed in the top floor of some high building capable 
of carrying a number of wires which radiated to the offices of users 
direct or were carried in groups to a more distant roof whence 
distribution commenced. There were thus two clearly defined 
types of line construction ' routes ' and ' distribution.' The 
routes were composed of a number of wires carried sometimes for 
considerable distances, whilst the distribution was from a defined 
centre over a given area. For routes as well as distribution, over- 
house construction was first adopted both in America and Europe. 
In the latter they were very generally so continued, but in the former 
facilities were afforded by the local authorities which permitted the 
routes to be transferred in large measure to pole lines in the streets, 
overhouse construction being retained principally for distribution. 

The increasing number of lines on routes and the difficulty of 
providing accommodation for them, led to cables being suggested as 
a source of relief. 

Under date of April 19, 1878, Mr. J. H. McClure, the manager of 
the Telephone Co., Ltd., ' sole proprietors of Bell's Telephone,' wrote 
a letter to Sir William Preece asking his opinion of the best form of 
cable for telephonic purposes, adding 

It is proposed to lay a cable from the various docks in London to a 
central office which, as you are aware, would be connected with 
various merchants' offices in the neighbourhood of such central 
office. I am of opinion that it would be less expensive and more 
convenient to lay a cable containing ten metallic circuits, than to lay 


a number of wires, if indeed we could obtain permission from the 
railway company for the larger number of wires : but I shall be 
glad to have your opinion on this point also. 1 

At this time Professor Bell was in London, and less than a month 
previously had issued the circular 2 to the Directors of the Telephone 
Company in which the use of cables is suggested and the con- 
nection with the docks referred to. 

One of the earliest instances of the use of cables as a means of 
bringing in a route of lines to a telephone central office was that 
at San Francisco in the year 1 879.3 A forty- wire cable 75 feet 
in length was first run from the roof into the exchange room, and 
later another thousand feet, in lengths of from 75 to 200 feet, 
were added. In 1881 there were some 7000 feet of different 
lengths, varying from 60 feet to 650 feet. " In that way [said 
Mr. Sabin] we manage to get into the office very well." 

The composition of the outside construction in New York as 
at May i, 1880, is recorded in an inventory which formed the basis 
of the valuation of the Western Union and Bell exchanges on that 
date. According to this inventory the Metropolitan Telephone and 
Telegraph Company (which was the title of the new organisation) 
received from the former companies 1892 miles pole wire, 970 miles 
house top wire, 1117 house top frame fixtures and from 70,000 to 
80,000 feet of cable averaging ten conductors each ; of this 25,000 
feet was aerial cable located principally over the East River Bridge 
to Brooklyn. 4 

The new Company 

constructed large and substantial lines to connect suburban territory, 
with a capacity of about fifty wires but carrying an average of about 
eight. On the top pin is strung a number 9 galvanised iron wire 
which is called the induction wire, and connected with earth at 
suitable intervals, also with metallic cups in the cross arms in which 
the insulator pin is inserted. 6 

As early as 1880 it was evident to those engaged in telephone 
work that the difficulties of construction and maintenance would 
necessitate a change from overhead lines. This was clearly stated by 
Mr. Pope at the first meeting of the National Telephone Exchange 
Association in the following words: 

1 I am indebted to Mr. A. H. Preece for this letter. 
* P. 89. 

8 Mr. John J. Sabin, National Telephone Exchange Association Report, 

4 Mr. H. W. Pope, National Telephone Exchange Association Report, 1880, 

P- 154- 

6 Ibid. p. 155. 


In New York and perhaps in Chicago the matter of underground 
lines must be soon taken up. Before coming to this convention 
that was made a prominent feature and was brought to my notice, 
and I was requested to do what I could to get a competent com- 
mittee on this subject. It should be a Committee who have had 
some little experience, or some knowledge that they have obtained 
abroad or elsewhere, in the matter of underground telegraphy. 
It is a subject that necessitates perhaps the establishment or the 
organisation of companies for this purpose, and it is a matter that 
needs immediate attention, and the thing should be gotten into 
shape. 1 

This statement was undoubtedly inspired by the controlling 
spirits of the business. The needs were becoming obvious, the 
difficulties foreseen, and the necessity for careful examination 
recognised. Some time was yet to elapse before an organised system 
of underground work was practicable. The cables then available 
were described and discussed. After describing one such system, 
Mr. E. M. Barton said : 

There is an important distinction in regard to the difficulties of 
laying subterranean and submarine cables and those of laying over- 
head cables, because there is not the difficulty of keeping out moisture 
in a cable suspended in the air or over buildings that there is under 
ground ; and while the question of making good cables is a 
prominent one for telephone managers to consider, the question of 
whether these cables shall go overhead or underground is also an 
important question. 2 

The question was not settled at that meeting. There was a con- 
current development along both lines ; but when it is realised that 
to bury the wires involved the use of a suitable cable and the 
excavation of the ground as well, whilst the use of aerial cable saved 
the latter expense, it is obvious, as Mr. Barton foresaw, that the aerial 
cable afforded the easier method of providing the relief which 
was becoming more and more pressing from the congestion of the 
wires on routes and especially so in the neighbourhood of the central 

The immediate problem was attacked in the light of the existing 
conditions. The eventual need of undergrounding was recognised, 
but it was clearly seen that a general underground system was a 
vast undertaking which required the most careful consideration 
and much more experience than was then available. Inquiry 
and the acquisition of experience on a smaller and safer scale were 
diligently pursued with a view to greater developments? 

1 Mr. H. \V. Pope, National Telephone Exchange Association Report, 1880, 
p. M7- * Ibid. p. 54. 


There was of course no novelty in either underground work or 
the suspension of aerial cables. 

Sir Charles Wheatstone, in his patent of October 10, 1860, con- 
templated the Use of a large number of electrical conductors in one 
cable suspended over housetops or on poles. And in the circular 
or prospectus of the Universal Private Telegraph Co., formed in 
1861, it is said : 

In place of the thick iron conducting wire in ordinary use fine 
copper wire (number 22 gauge) is employed, carefully insulated 
over its entire length with a thin coating of india-rubber by Messrs. 
S. W. Silver & Co.'s patented process, and further protected from 
abrasion by a covering of tarred tape. Thus perfectly insulated 
from one another, these ' electric highways ' are afterwards com- 
bined into ropes of about two hundred and thirty yards in length, 
containing twenty, thirty, fifty or one hundred wires according to 
the requirements of the district through which they are intended 
to pass. 

Mr. Lockwood stated in 1882 that : 

Until an extensive network of telephone lines had been erected, 
very little aerial cable construction had been done in the United 
States. When the Gold and Stock Telegraph Company moved its 
offices and operating room from 61 to 195 Broadway in New York, 
it was requisite to retain the former place as a testing station ; 
and all the wires were accordingly laid from 195 to 61, before pro- 
ceeding out to the brokers' offices. To accomplish this, they were 
carried in kerite cables of nine and ten wires each. This was in 
1874. In 1875, the Law Telegraph Company was organised, and 
also to a certain extent adopted cable construction. 1 

The aerial cables for telegraphic use were almost invariably 
composed of wires separately insulated with india-rubber or 
some substitute therefor, such as the kerite above mentioned. 
The use of aerial lines, whether suspended singly or in cables, was 
adopted only after the experience of unsatisfactory results from 
underground lines, for on its introduction it was contemplated 
that the telegraph would be used through the medium of buried 

The demonstration line laid by Sir Francis Ronalds in the garden 
of his house at Hammersmith in 1816 (later the Kelmscott House of 
William Morris) was an underground line. A trough of wood two 
inches square, well lined both inside and out with pitch, was laid in a 
trench four*feet deep. In the trough were placed thick glass tubes 
jointed with soft wax, and the wires were run through the glass tubes. 

1 National Telephone Exchange Association Report, 1882, p. 90. 


Mr. Lockwood pays a tribute to the prescience of Sir Francis 
Ronalds, who 

described methods of insulating the wires, either on poles or 
underground, with all the details of tubes, joints and testing boxes, 
testing stations, linemen and inspectors, as at the present day. 
But the most wonderful feature of the affair is that Ronalds clearly 
foresaw the great trouble which, even at the present day, limits the 
speed of submarine and subterranean telegraphy, and apparently 
bars the way to successful telephony on underground lines of great 
length, i.e. the retardation of signals, due to the induction of the 
earth ; for he says in his book, which was published in 1823, ' That 
while over his short line the charges were apparently instantaneous, 
he did not contend, nor even admit, that an instantaneous discharge 
through a wire of unlimited extent would occur in all cases.' 
Further on he says : ' That objection which has seemed, to most of 
those with whom I have conversed on the subject, the least obvious 
seems to me the most important, and therefore I begin with it, 
viz., the probability that the electrical induction which would take 
place in a wire enclosed in a glass tube of many miles in length (the 
wire acting like the interior coating of a battery), might amount to 
the retention of a charge, or at least might destroy the suddenness 
of the discharge, or, in other words, might arrive at such a degree 
as to retain the charge with more or less force, even when the wire 
is brought into contact with the earth.' 1 

And Sir William Preece said of Ronalds : ' It is perfectly 
astonishing how that man's instinct saw the various troubles that were 
likely to be met with in the construction of long underground lines.' 2 

Messrs. Cooke and Wheatstone's 1837 patent included a plan for 
laying subterranean wires, and in that year they established tele- 
graphic communication between Euston Square Station and Camden 
Town, the conductors being carried underground. The means adopted 
in this case were to place five copper wires, each separately covered 
with cotton and insulated by a preparation of resin, into five grooves 
cut longitudinally in a piece of timber. Tongues of wood were placed 
over the grooves, the whole covered with pitch and buried in the 
ground. Thus in Great Britain the first telegraph lines were laid 
underground, and it was intended that the first in the United States 
should also be so laid. Ten miles of the line from Baltimore to the 
Relay House were laid underground in December 1843. The line 
consisted of four No. 16 copper wires covered with cotton and 
shellac drawn into lead pipes. This experiment was a complete 
failure, and the wires were taken out of the tubes and placed on 
poles. The remainder of the line was also placed upon poles. 

1 National Telephone Association Report, 1882, p. 80. 
* Journal of the Society of Telegraph Engineers, xvi. 427. 


It may be presumed that the method adopted in 1837 had by 
1839 been proved to be unsatisfactory, for in the latter year the 
wires on the Great Western Railway were carried as far as \Yest 
Drayton through an iron tube an inch and a half in diameter, fixed 
about six inches above the ground, parallel to the railway and 
about two or three feet distant from it. Chambers' s Journal (1840), 
from which this information is taken, further states that : 

Each of the four wires in Professor Wheatstone's apparatus 
is wrapped round with a well-rosined thread, and the whole are 
then tied together with a cord possessing a similar coating, so as 
to present the appearance of a tightly bound rope. This it is 
proposed to place in a small iron tube like that used for bringing 
gas into houses, and the tubes, united to any length, are laid below 
the ground, or in a wooden case on the surface, to preserve them 
from injury. 1 

These early efforts were succeeded by others which are concisely 
recorded by Mr. Fleetwood in his paper read before the Society of 
Telegraph Engineers (I.E.E.) in 1887,2 to which reference may be 
made for details. Up to 1849 the insulation was resin ; a mixture 
of tar, resin, and grease ; or some similar compound. Conductors 
(generally four) covered with cotton or yarn and so insulated were 
drawn into lead tubes about half an inch diameter. Two such 
tubes were drawn into a three-inch cast iron socket pipe (still the 
standard size), thus allowing eight wires to the three inches. Lines 
of such a type were laid in 1846 from 345 Strand, the office of the 
Electric Telegraph Co., to Nine Elms Station, the then metropolitan 
terminus of the London and South-Western Railway. After 1848 
gutta-percha took the place of the compounds previously used and 
the wires were drawn directly into the iron tubes. 

From 1850 to 1854 there was great activity in telegraphy, and 
underground wires were laid not only in cities but for long-distance 
work such as London to Dover (six wires), London to Liverpool 
(six wires by one company and ten by another). But the life of 
these lines was short, for in 1857-8 they were condemned and lines 
on poles substituted. Mr. Fleetwood considers the failure to have 
been due to defective covering on some of the wires and still more 
to defects in the process of laying. 

The chief cause of the rapid decay of the lines was the rapidity 
with which the work was carried out and the absence of supervision. 
The majority of the men employed were totally ignorant of the 
necessity for the greatest care being exercised, and it was impossible 
for the few able men to be everywhere along the works at the same 

1 Chambers's Edinburgh Journal, July 25, 1840, ix. 209. 
* Journal of the Society of Telegraph Engineers, xvi. 404. 


time, so that it was not surprising that these early lines had to be 
so soon abandoned. 

Mr. Robert Sabine, reporting on telegraphs in the Paris exhibition 
of 1867, compares the six underground systems in that exhibition 
with the one only shown in the London exhibition of 1862, 

plainly indicating the increased attention which is being devoted 
to this [underground] branch of telegraphy, and therefore a growing 
necessity for it. ... Underground lines, almost abandoned in 1862, 
have since that date been creeping gradually again into favour. 1 

Mr. Sabine, writing at a period nearer than that of Mr. 
Fleetwood, gives the same reason for the earlier failures. ' The 
principal difficulties,' he says, 

which underground lines have had to contend with have been the 
carelessness with which the wires were laid and the decay of the 
insulating materials. These difficulties have been met more com- 
pletely in France than elsewhere. 1 

He describes some of the exhibits in which gas tar, asphalte, 
or bituminous compound of some sort are the insulating substances. 
One system (Donald Nicoll, Kilburn) had an ingenious method of 
jointing the conductors of the separate sections. At one end the 
wires were twisted in hollow coils ; at the other they were straight 
and the straight end was pushed into the adjoining helix and 
soldered. A great point in favour of the proposed system in Mr. 
Sabine's opinion was that it required little skill or practice on the 
part of the workmen employed. Another exhibit (A. Holzmann, 
Amsterdam) was a trough system in which the wires were separated 
by glass supports, which Mr. Sabine considered objectionable. 
But of both he says : 

These systems are not new : they have been tried repeatedly 
in France and in England, but failed, partly because the material 
employed for the insulation was too brittle, partly because insuffi- 
cent care was taken to keep the wires apart. But with the benefit 
of all the experience and failures of other inventors, it is to be 
hoped that both Mr. Nicoll and Mr. Holzmann may succeed in their 
endeavour to give us cheap underground lines, by which they will 
be doing a most welcome service to telegraphy. 1 

It is evident that this hope was not realised, for the draw-in 
system became general. The reason may be gathered from Culley, 
who says that it is very much more easy to place the wires in a 

1 Illustrated London News, November 16, 1867, p. 550. 


trough than to draw them into a pipe, but it is much more difficult 
to execute repairs in a trough. 

The plan which has been pursued for many years in London and 
the larger towns is to lay down a pipe under the flagstones amply 
large enough for all the wires likely to be required. Oblong ' draw- 
ing-in ' boxes, 30 inches by n inches and 12 inches deep . . . are 
placed at every hundred yards if the line be straight, and nearer 
if it be curved. 1 

On the transfer of the British Telegraphs to the State in 1870, 
considerable additions to the underground work in London were 
made ; and on January 17, 1874, when the wires were diverted to the 
new office in St. Martin's le Grand, the length of the pipes in the 
metropolitan district was about 100 miles, and the total quantity of 
wire in the pipes amounted to 3000 miles. Later, a new line of 
three-inch cast-iron pipes was laid down and 100 miles of wire 
drawn into it, in substitution for 126 miles of overhouse wire. 2 A 
route of underground lines was also laid between Liverpool and 
Manchester by the Post Office in 1870. 

The interest of the citizens and the Government in the amenities 
of the capital city led to the underground system in Paris being 
developed to a considerable extent, and in 1864 it was well reported 
on by Professor Hughes. There were (he said) over 2,000 kilometres 
of wire laid, all in a perfect state of high insulation, which, from the 
constant and uniform good results, made the French system worthy 
to be studied by electrical engineers. There were seven distinct 
underground routes leading from the central station in Paris to the 
limits of the city, where they were joined to the air lines leading to 
the different provinces of France. Iron tubes 2\ yards long were 
laid in a trench one yard deep. A pilot wire was placed in the tubes 
whilst being laid. With this a rope was drawn in serving to 
introduce the cable. He adds : 

There are now 52 kilometres of subterranean lines in Paris, 
1450 kilometres of wire enclosed in these tubes. . . . Tubes of 
60 millimetres receive easily 28 wires ; 100 millimetres receive 49 
wires ; those of 120 millimetres have 77 wires. The copper-con- 
ducting wire is formed of a strand of four wires, each having a 
diameter of half a millimetre. They are covered with two coverings 

1 A Handbook of Practical Telegraphy, fifth edition, 1871, p. 153. 

* The process of placing the wires underground wherever possible has been 
considerably extended, and of the 1,745 lines of wire entering the Central 
[Telegraph] Station in London, not one is open. In many cases the wires 
are conducted underground for distances of 12 to 22 mile? from this office. 
(Postmaster-General's Report, 1888, p. 9.) 


of gutta-percha, giving in total a diameter of five millimetres. These 
are separately covered with tarred hemp. 1 

Professor Hughes gives a record of his tests, and he attributes 
the satisfactory results attained to the fact of the iron tubes being 
solid and hermetically sealed, the wires being thus protected against 
any mechanical injury as well as from the effects of gases which were 
considered to be the main cause of the deterioration of the gutta- 
percha. He gives also particulars of the capital outlay which will 
be of interest in comparison with that for existing underground 
circuits : 

Iron tube, 120 millimetres diameter, in which are placed sixty- 
three wires, contained in nine cables of seven wires each. The 
entire cost of the tube, trenches, placing wires, etc., complete, 
eight francs per yard, or twelve centimetres each wire per yard. 
The cables, with seven insulated wires, cost 2900 francs per kilo- 
metre. This would make 414 francs each wire per kilometre, and, 
adding the cost of the iron tube, with laying, etc., the total price 
would be 534 francs per kilometre ; and as these tubes are, after 
four years' use, in a most perfect state of preservation, it is but 
reasonable to suppose that they will be serviceable for many years 
to come. 1 

While in Great Britain and the United States the system adopted 
was determined by conditions of economy or commercial efficiency, 
the exigencies of military requirements had some sway on the 
Continent and probably account for the further efforts at the 
development of underground work in both France and Germany. 

The later German system of underground telegraphs dates from 
about 1876. The cables were made on the general plan of an 
ordinary submarine telegraph cable. Each cable contained seven 
conductors. Over 2,000 miles of cable, containing 15,000 miles 
of wire, connected the principal cities in 1882. 

In the United States, underground telegraphs were not exten- 
sively used, but when adopted the European practice of separately 
insulated wires with suitable protection by pipes or wooden con- 
duits was followed. 

In 1875 William Mackintosh, Superintendent of Construction 
for the Western Union Telegraph Company, laid a cable underground 
through a three-inch pipe from Broadway and Dey Street to 
18 Broad Street, New York. It was composed of five conductors 
of No. 16 copper. 

The insulation was ' best English gutta-percha.' The length 
of this underground line was about 1600 feet. The iron pipes were 

1 Telegraphic Journal, No. 33, August 13, 1864, vol. ii. p. 73. 


dipped in coal tar and the joints were leaded. The cable cost 
2\ cents per foot, the iron pipe about 50 cents per lineal foot, and 
the digging, laying, and repaying a further 50 cents per lineal 

Since the laying of this cable, another of twenty-eight wires has 
been added, and was pulled through the pipe a distance of 800 feet 
by the aid of a team of horses and twenty-five men. This under- 
ground line was opened and inspected about four months ago 
(December 1880) and found to be in perfect condition and the gutta- 
percha as pliable as ever. 1 

The early underground work, in which the insulation of numerous 
conductors was dependent upon one outer covering, had failed and 
the causes of its failure are attributed in part to the materials and 
in part to the want of care or the impracticability of obtaining the 
necessary conditions to do satisfactory work in the process of laying. 
But efforts were made to overcome these difficulties, and it is some- 
what remarkable that amongst the earliest efforts are to be found 
the principles of the latest successes. On August 4, 1845, William 
Young of Paisley, manufacturer and dyer, and Archibald McNair 
of the same town, merchant, obtained a British patent (No. 10,799) 
for ' an improved method of manufacturing electrical conductors.' 
The conductors were to be ' formed of one or more copper, iron, or 
other metallic, or mixed metallic, wires.' The preferred method 
of covering the wires was ' with threads in a plaited or braided 
form by means of a braiding machine,' stress being laid on the 
suitability of the covering thus applied to withstand the subsequent 
processes of manufacture. The wires, having been covered, were 
wound on reels to which was applied suitable tension. They were 
unwound from the reels through a lead press, first passing through 
a cistern of molten pitch on their way through the press. Two 
methods of lead-press operation are described ; ' containers ' and 
' die blocks ' are fully illustrated ; the consistency and temperature 
of the lead defined, and suitable methods of applying heat to both 
the insulating material and the lead suggested. After passing 
through the lead press the cables were to be wound on reels, the 
reels mounted on wheeled carriages, and, one end of the cable being 
held fast, the carriage was to be driven in the desired direction, the 
cable unwound and delivered on the ground ' without risk of injury 
and with great facility.' Having been laid, the ends were to be 
brought into joint or test boxes provided with mercury cups through 
which the continuity of each conductor and of the sheath was 

The scope of the claim would indicate that this may be the first 

1 National Telephone Exchange Association Report, April 1881, pp. 49-50. 


recorded suggestion of such a method of manufacture, for while the 
patentees say they have no intention to claim 

the adaptation of wires surrounded with non-conducting substances 
enclosed in tubes for electrical conductors, . . . that which [they] do 
claim is, the construction and manufacture of the electrical conduc- 
tors by the employment of machinery having a tubular mandril 
or hollow rod through which wires may be drawn, whilst the leaden 
or other soft metal tube is forming, by pressure between a core and 
die, such wires being at the same time imbedded in pitch or other 
non-conducting material. 1 

The patent specification contemplates a multi-conductor cable, 
but if ever made by this process such a cable does not appear to 
have reached a commercial stage. A single conductor insulated 
with gutta-percha and enclosed in a leaden tube was exhibited by 
McNair in the London 1851 Exhibition. 2 According to the Jury 
Report there were many specimens of subaqueous wire in the 
Exhibition, but McNair 's wire was the basis of most of them. The 
process of manufacture is briefly but clearly described in the Report, 3 
and the opinion is expressed that the article seems to answer its 
purpose well. 

The adoption of a similar process of covering in the United 
States is shown in Shafmer's statement 4 that ' Mr. Bishop has 
constructed extensive machinery for the covering of the insulated 
wire with lead of any required thickness.' 

Enclosing a single conductor the hydraulically pressed lead 
covering was an article of commerce in 1851, but for a multi- 
conductor cable it is probable that there was not sufficient con- 
fidence in the lead covering so formed, for in 1869 William Alfred 
Marshall of Canonbury, in the County of Middlesex, telegraphic 
engineer, took out a patent 5 ' for improvements in the process of 
insulating and protecting the wires of subterranean, submarine, 
and other electric telegraph cables.' The wires were to be covered 
with cotton or other fibrous non-conducting material, which was 
to be previously dried and then placed in a vessel containing 
melted paraffin wax until the whole of the cotton or other fibres 
were thoroughly permeated. The wire or wires thus covered 
or prepared were introduced into a lead or other soft metal tube 
of convenient length laid or stretched horizontally in a straight 
line. The lead tube was to be wound on a skeleton drum to which 

1 British specification, No. 10,799, 1845. 

2 Official Description and Illustrated Catalogue, i. 455. 

3 Reports of the Juries, p. 293. 

* Telegraph Manual, 1859, p. 606. 
, December n, 1869. 


heat was applied. At one side of the axle of the drum was a tube 
extending through the bearings. The inner end of the tube ter- 
minated in a nozzle to which one end of the cable's leaden covering 
was attached. The outer end of the tube had also a nozzle to which 
was connected a flexible tube attached to a pump. The air was 
to be first pumped out of the cable and paraffin was then pumped 
in. The full description of the process, which is here abbreviated, 
shows that Marshall had a clear idea of what he required to attain 
in the course of manufacture. The reason for first expelling the 
air is thus stated : 

By introducing the paraffin wax into the tube [the cable covering] 
in which a vacuum is created as above explained, I prevent the 
formation of air bubbles therein, which would impair the insulating 
properties of the paraffin wax. 

And further 

In order to provide for the contraction of the paraffin wax on 
cooling, I proceed to cool the cable gradually in the direction of its 
length, commencing at the disengaged end. 1 

Recognising that there was no novelty in the use of cotton, 
paraffin, or lead or in their combination, Marshall, like the patentees 
previously quoted, made ' no claim to the use of any of the materials 
mentioned,' but limited his claims to (i) the improved process 
described ; (2) the skeleton drum for carrying the length of cable 
in combination with the heating tank (or hot air chamber) ; and 
(3) the combination with the skeleton drum of the hollow journal, 
the coil and tube for the supply of paraffin wax from tank to cable 
during the rotation of the drum. 2 

A length of this cable was laid across Windsor Park (in a clay 
soil) in connection with a private telegraph line, but, after being 
down for some months, it failed, and on examination it is said the 
lead covering was found to have become completely decayed in 
several places, thus letting in moisture and destroying the 
insulation. 3 The cause of the decay in the lead pipe appears to 
have been attributed to the clay soil in which the cable was laid, 
but experience has shown this to be inaccurate. Decayed vegetable 
matter in the soil may, however, have been the destructive agent. 

Marshall's patent appears to have attracted but little attention, 
and a single failure seems to have been sufficient to close its com- 
mercial career. It deserved a better fate. Even with the defects 
which are noticeable, more extended trial would have shown it to 

1 British specification, No. 2587, 1869, p. 6. 

* Ibid. p. 9. 

8 Electrical Review, London, November n, 1882, p. 368. 


be reliable. It was an early attempt to transfer to a factory the 
combination in a completed form of conductors, insulation, and 
covering, and the attempt was made in a manner which (but for 
the slight yet important feature of the air treatment) led to eventual 

Confidence in the outer covering was essential for the com- 
mercial success of any system in which the insulation of the interior 
wires was dependent upon the integrity of the envelope, and the 
experience with lead-covered cables was not encouraging. David 
Brooks of Philadelphia was of opinion that the lead pipe was 
liable to damage in the process of coiling and uncoiling on the 
drum, and he proposed the use of iron pipes with a liquid insulation. 
His system, patented in the United States July 13, 1875,* and 
Great Britain (No. 4824) December 19, 1877, had the merit of 
simplicity. Cotton-covered wires were drawn into iron pipes 
of which the joints were thoroughly caulked. The pipe was then 
filled with mineral oil under pressure, so that, in the event of any 
defect in joint or pipe, the tendency should be for the oil to leak 
out rather than for the moisture to leak in. Introduced in the United 
States in 1877 or 1878, the United States patents on the Brooks' 
system were acquired by General Anson Stager on behalf of the 
Western Electric Manufacturing Company in i879- 2 Mr. Barton 
stated in 1880 3 that some of the cables laid at an early period were 
still in operation. Between twenty-five and fifty sections of different 
lengths had been used for telephone lines for a period of a year to a 
year and a half. Whilst some had failed from mechanical injury, 
the others had been successful. A pipe an inch and a quarter in 
diameter was provided for a core of fifty wires. 

The system was introduced into England and Belgium. 
An illustrated description appeared in the London Telegraphic 
Journal of December i, 1879, and in 1880 Mr. Brooks made a 
proposition to the Postal Telegraph authorities that an experi- 
mental length should be laid between Waterloo and Nine Elms on 
the London and South- Western Railway, on the condition that it 
should be paid for if it worked satisfactorily for a period of six 
months. 4 The proposition was accepted, the line laid, and the 
conditions met. In the following year (1881) the line was extended 
to Clapham Junction. The extension was carried out by the India 
Rubber, Gutta Percha and Telegraph Works Company under the 
direction of Mr. Brooks. 5 It was completed during the summer, 

1 U.S. specification, No. 165,135. 

2 Journal of the Telegraph, (New York,) May 16, 1879, p. 149. 

8 National Telephone Exchange Association Reports, 1880, p. 47. 
4 Telegraphic Journal, August 15, 1881, p. 306. 

8 Some personal recollections of the laying of this line are given by 'J.T.L.' 
in the Post Office Electrical Engineers Journal for July 1915, viii. p. 118. 


and tests published in the Telegraphic Journal. The use of this 
line for telephone purposes is thus stated : 

Experiment shows that the Brooks' system may be employed with 
great advantage for telephone working ; a single wire of the cable 
between Waterloo and Nine Elms has been employed as a telephone 
circuit for some time, and, although the other twenty-nine wires 
are heavily worked, yet the inductive interference has not been such 
as to cause inconvenience. A telephone, placed on two wires of the 
entire length between Waterloo and Clapham did not emit any 
sound, although a Wheatstone automatic instrument with high 
battery power was being worked on one of the other wires ; the 
metallic loop, it 'may be observed, was not a twisted one, but was 
formed of two wires picked at random from the group of thirty. 
No doubt the satisfactory result obtained is in a great measure due 
to the fact that the relative positions of the various circuits in the 
pipe is different at every point throughout the length. In the cases 
where a number of wires in a cable are employed for the telephone 
working alone, it is found that the inductive interference between 
wire and wire diminishes in proportion to the number of wires in the 
pipe, a result due, no doubt, to the distribution of the effect. 1 

This experimental line continued to work well up to i882, 2 
but the eventual result is thus told by Mr. Fleetwood : 

After this line had been completed, the wires, with one or two 
exceptions, gave very good results as regards insulation ; but from 
the first there was a very great loss of oil, and this continued more 
or less although every effort was made to trace and repair the 
leaks. Probably the constant vibration of the viaduct has had 
something to do with the leaky condition of the joints. Owing 
to the viaduct being widened for a considerable length, it was 
necessary last year to divert the wires working through this system 
between Westminster Bridge Road and Queen's Road Station, and 
to recover the pipes and wires along the railway between the above 
points. This was done, and the remaining portion from Queen's 
Road to Clapham Junction thoroughly overhauled, one section of 
about 300 yards being renewed. Since last October the line has 
worked well, and, although the cost of maintenance has been very 
heavy, it does appear to me that the system is worth a further trial 
under more favourable conditions for instance, along a country 
road, where it might form 'part of a through line and where the 
pipes would not be subjected to such frequent disturbances as in 
the busy thoroughfares of London. 3 

The liquid oil system of insulation was, however, found to be 

1 Telegraphic Journal, August 15, 1881, ix. 306. 

2 Electrical Review, London, September 30, 1882, xi. 262. 

3 Journal of the Institution of Electrical Engineers, xvi. 418. 


unsatisfactory even on routes not subjected to vibration, some- 
times from leakages, sometimes possibly from the absorption of 
moisture. It had, moreover, the defect of previous systems in that 
it was not complete until it was laid. It was not an article of 
manufacture but a combination of conduit and cable separately 
constructed and the combination completed in situ. It did not 
survive, but it deserves commendation and remembrance as the 
starting-point from which subsequent developments arose. 

The Brooks cables were laid in the United States under the 
supervision of Mr. W. R. Patterson. His company sought to 
improve the existing types of multi-conductor cable, and, in common 
with the rest of the world, they had the knowledge from experience 
that liquid oil in a rigid pipe would leak out and solid wax in a 
flexible pipe would crack. Liquid oil in the flexible pipe or solid wax 
in the rigid pipe were considered to be equally impracticable. 

On May 14, 1881, Mr. Patterson filed an application for a patent, 
in which he related the facts that paraffin, resin, beeswax and other 
similar substances alone or in combination had been used as an 
insulating substance, and it was weh 1 known they shrink on becoming 
cold ; thus cavities had been left within the pipe which had caused 
great trouble. 

In order to compensate for this shrinkage [said Mr. Patterson] 
I charge the melted paraffin or other insulating substance with 
carbonic acid gas or other suitable gas or mixture of gases and force 
the liquid thus prepared into the heated pipe around the core. . . . 
The gas is held in very minute bubbles hardly visible to the naked 
eye. As the substance cools and contracts, these bubbles expand 
and the pipe is thus kept entirely full. These bubbles are isolated 
from each other and uniformly diffused throughout the mass of solid 
substance. The insulating substance is thus formed into a light, 
porous, homogeneous mass. 1 

Patterson's cable was Marshall's cable with one very important 
difference. Marshall excluded air because he conceived it would 
be detrimental to the insulation, and Patterson included air in the 
form of a suitable gas because it would render the insulating sub- 
stance with which it was combined elastic instead of brittle, 
expansive instead of contractive. 

The manufacture of this type was immediately undertaken. 

At the meeting of the National Telephone Exchange Association 
in 1883 it was stated by Mr. Fay of Chicago : 

In 1882 a loo-conductor Patterson lead-pipe cable was placed 
in the Washington Street tunnel, which is still working and in good 

1 U.S. specification, No. 248,209, October n, 1881. 


Some time prior to this, the system of housetop wires had grown 
into a forest of tall frames, cumbrous to handle and obnoxious to 
property owners. In the spring of 1882 it was decided to commence 
a gradual but sweeping change in this city by substituting aerial 
cables wherever the bodies of wires exceeded forty in number. 

It is not improbable that, at some time, city exchanges will be 
operated generally upon the metallic circuit system ; and in buying 
cables which were intended to last for some years, it seemed advis- 
able to have them made of pairs of conductors twisted together, so 
that each pair could be used as a metallic circuit, if desired, in the 
future. We purchased some Patterson lead cables, made by the 
Western Electric Company, with this in view. They consisted of 
fifty pairs of conductors of No. 26 copper wire. One wire in each 
pair is insulated with white cotton and one with red. 1 

A catalogue of the Western Electric Manufacturing Company 
"xlated 1882 gives a list of Patterson cables then in practical 
operation (see table on p. 257). 

The aeration of the paraffin wax resulted not only in the pro- 
duction of an elastic insulating material well adapted as a filling 
for cables, but was also of great benefit in another respect. The 
advantages of low capacity were well recognised by telegraph 
engineers, and David Brooks had included this amongst the advan- 
tages of his liquid oil system, but the Patterson system was a further 
advance. A descriptive circular of 1884 states this advantage as 
follows : 

Another and not less important function of the gas is to reduce 
the specific inductive capacity of the insulator. Solid paraffin 
possesses about the lowest static capacity of any insulator, except 
air and gases ; an admixture of dry gas in the above-described 
manner reduces the specific inductive capacity at least 15 per 
cent, below that of pure solid paraffin. 

The practical utility of this extremely low inductive capacity is 
that it enables us to use a smaller conductor than if such materials 
as gutta-percha or rubber were used as insulators. Large conductors 
will give better results than small ones, other things being equal. 

But as the retardation is affected equally by the capacity and 
resistance of the wire, if we reduce the specific inductive capacity 
of the insulating covering, the resistance of the conductor can be 
increased in the same ratio without affecting the working of the 
circuit ; or, to state it in another way, if the same size of wires is 
used, the cable with the smallest capacity will give service through 
the longest distance. Conversation can be carried on with equal 
loudness and clearness through 1500 ohms and 10 micro-farads 
as through 1000 ohms and 15 micro-farads. 

The use of small conductors is advantageous in some respects 
provided the retardation is no greater than in large ones. 

1 National Telephone Exchange Association Report, 1883, p. 68. 



The lead-tube type of cable was generally adopted for both 
underground and aerial purposes, and its manufacture was now 
undertaken by several firms. The first to be put into actual use 

Subterranean and Subaqueous 


Number of 

Size of Con- 





Evansville, Ind. 








Norfolk, Va. 




Norfolk, Va. 




McGregor, Iowa. 




McGregor, Iowa. 




Bridgeport, Conn. 




Buffalo, N.Y. 




Newburyport, Mass. 




Ludington, Mich. 




Green Bay, Wis. 




Michigan City, Ind. 




Toledo, Ohio. 




St. Joseph, Mich. 




Trenton, N.J. 

7 8 




A erial 


Number of 

Size of Con- 

















40 26 

Jersey City. 

[The size of conductors above given is that of the B and S gauge.] 

for telephone purposes would appear to have been the Philadelphia 
cable of Brooks Junior mentioned by Mr. Sargent at the meeting 
of the National Telephone Exchange Association in April iSSi. 1 
Mr. Sargent stated that they had a cable in Philadelphia which 
was laid some time in May 1880 and had been in successful operation 
ever since. The length was about 600 feet, and there were eighty-four 

1 National Telephone Exchange Association Report, April 1881, p, 38, 


twin wires in an inch lead pipe. It was made by David Brooks 
Junior a son of the inventor of the Brooks cable previously described. 
The insulation was a combination of resin and coal oil forced in 
in a warm state and becoming solid when cold. In the manner of 
making splices and in the material employed, this cable was similar 
to that used by the Electric Telegraph Company between the Strand 
and Nine Elms Station in London in 1846. It differed by multiplying 
more than tenfold the number of wires within one lead covering 
and by transferring to a factory much of the work hitherto^ done 
upon the line, just as Young and McNair and Marshall had proposed, 
the only difference being in the material used as insulation. 

The Patterson cable was the next of this type 
put into commercial use, and was followed by the 
Waring System of the Standard Underground Cable 
Company. In a descriptive circular of the Waring 
Underground Electric System (apparently issued in 
1887) it is said that Mr. Waring's system had been 
developed and applied ' during the past five years.' 
The insulation is described as a hydro-carbon com- 
pound, inorganic in its composition, not liable to 
disintegration or deterioration. . . . Unlike paraffin, 
gutta-percha, and many other commonly used 
dielectrics, the Waring insulation may be subjected 
to a ' high degree of temperature without decen- 
tralising, carbonising, or breaking up into vapours 
and gases, but upon cooling returns to its normal 
condition, unimpaired in its insulating qualities.' 
The cables are classified as follows : 

FIG. 104. 
Waring Cable 

1. Anti-Induction Cables. Each insulated conductor being 
separately surrounded by a body of metal, thus destroying the effects 
of induction. Used for telephone purposes, where secrecy is impor- 
tant and distinct articulation a prime requirement ; for telegraph 
and general electric purposes, where exactness of current, free from 
extraneous influences, is necessary to the proper adjustment and 
working of instruments. (Fig 104.) 

2. Bunched Cables. Where a greater number of conductors is 
required in a minimum space. Used for ordinary telegraphic, 
telephonic, and other electric purposes. (Fig 105.) 

4. Submarine Cables. Either anti-induction or bunched. Used 
for all above purposes where crossing water en route becomes 
necessary. (Fig. 106.) 

Amongst the earliest manufacturers in Europe to develop 
the lead-pipe type of cable were Messrs. Berthoud & Borel, of 
Cortaillod, Switzerland. In 1879 they introduced a lead-pipe cable 



which had also a leaden conductor. The latter was a peculiarity 
and was adopted with a view to the simultaneous manufacture of 
core and tube with insulating material separating them. 1 The 
high resistance of the lead conductor and the unstable nature of 
the insulator (sulphur or crushed resin) prevented this cable from 
being successful ; but in 1881 the makers introduced a modification, 
returning to copper as a conductor, using cotton as a separator and 
paraffin as an insulator, but retaining the method of putting on 

FIG. 105. Waring 
Cable (Bunched). 

FIG. 106. Waring 
Cable (Submarine). 

the lead covering in the process of manufacture. In this respect 
Messrs. Berthoud & Borel were, so far as practice was concerned, 
in advance of the times though, as we have seen, the process was 
patented in 1845. The Standard Underground (Waring) Co. also 
adopted lead press covering. The early Patterson cables were 
made by drawing a core of suitable length into already constructed 
lead pipes and jointing those pipes together. To quote the 
Patterson circular : ' The core may be made in continuous lengths 
of 1500 feet and the protecting pipe is jointed over it in lengths of 
75 to 100 feet.' The reason for this was the lack of confidence in 
the integrity of the lead pipe without a previous test. ' Any flaw 

1 Telegraphic Journal, April i, 1879, p. 116. 

S 2 


which may exist in the pipe and any leaky joint is detected by the 
process of filling.' 

Messrs. Berthoud & Borel, having made up the core, passed it 
through a lead press to receive its covering. The process is described 
in the London Electrical Review of November n, 1882, where 
it is stated that nearly ten miles of these cables were employed 
for the transmission of power and light in different parts of the 
[1881 Paris] exhibition building ; 5000 metres had since been laid 
down for the Societe Jablochkoff for the lighting of the opera ; a 
cable with several conductors, placed in the Paris sewers, connected 
the Ministry of Telegraphs with the exhibition, and several kilo- 
metres had also been laid down in the sewers for telephonic trans- 
mission. 1 The lead press was shown in operation at the exhibition. 2 

That there was some ground for the general lack of confidence 
then felt for a covering so made is suggested by the provision of a 
second lead covering when the cables were intended for under- 
ground use, an ' impermeable substance such as gas tar ' being 
placed between the two sheathings. 

Some few years later the employment of the hydraulic press 
system for lead covering was general, but, in the early stages of the 
efforts to supersede the separately insulated individual wires of 
the core by a group of wires dependent for their insulation on 
the mechanical envelope, confidence in the latter was of prime 
importance. The continued use of lead-covered cable prompted 
improvements in hydraulic lead presses which effected great economy 
in manufacture and left no doubt of the product, so that all manu- 
facturers eventually adopted the method which Young and McNair 
patented in 1845 and Messrs. Berthoud & Borel commercially 
revived in 1879. 

The characteristics of the telephonic current and the extreme 
sensitiveness of the telephone receiver differed so materially from 
the currents and the instruments used in telegraphy that a new 
line of experience was necessary in order to demonstrate the effect 
in practice of combining conductors together in cables and using 
those conductors for speech transmission in the form of individual 
messages for separate and distinct correspondents. 

There was considerable experience of the conditions governing 
the use of telegraphic currents, and the effect in practice was known 
so far as the transmission of detached signals at intervals was 
concerned. A Joint Committee was appointed by the Lords of the 
Committee of Privy Council for Trade and the Atlantic Telegraph 
Company to inquire into the construction of submarine telegraph 
cables. Their report was published in 1861. To this Committee 

1 Electrical Review, London, xi. 368. 
1 Mechanical World, December 30, 1882. 


Sir Charles Wheatstone submitted the results of experiments which 
were made by him to determine the amount of inductive discharge 
in wires of very considerable lengths according to variations in the 
lengths of the wires, their diameters, the material and thickness of 
the insulating substance, and in the temperature and pressure of 
the medium by which they were surrounded. The introductory 
paragraphs of Sir Charles Wheatstone's report will serve to illus- 
trate one of the effects incident to the use of cables applied to 
telegraphs : 

The introduction, in late years, of submarine telegraphic cables, 
and the substitution of subterranean lines of considerable lengths 
for the aerial lines almost exclusively employed formerly, have 
brought into evidence certain conditions of electrical charge which 
so materially influence the rapidity and frequency of the signals 
transmitted as to make it an object of the highest importance that 
they should be strictly investigated, in order that it may be ascer- 
tained whether it be possible to obviate the evil effects arising 
therefrom, or in any degree to alleviate them. 
1 These conditions could not fail to come under the observation of 
persons employed in telegraphic operations as soon as such lines 
were constructed ; but before they were made the subject of 
investigation by Mr. Werner Siemens 1 and Dr. Faraday, 2 only 
vague notions respecting their cause prevailed. 

When a metallic wire is enveloped by a coating of some insulating 
substance, as gutta-percha or india-rubber, and is then surrounded 
by water or damp earth, the system becomes exactly analogous to a 
Leyden jar or coated pane ; the insulating covering represents the 
glass, the copper wire the inner metallic coating, and the water or 
moist earth the external coating. The electricity with which the 
wire is charged, by bringing the pole of an active battery in contact 
with it, acts by induction on the opposite electricity of the sur- 
rounding medium, which in its turn reacts on the electricity of the 
wire, drawing more from the source, and a considerable accumula- 
tion is thereby occasioned which is greater in proportion to the 
thinness of the insulating covering. 

One mile of copper wire, one-sixteenth of an inch in diameter, 
presents a surface of 85-95 square feet, and receives the same 
charge from a source of the same tension as a Leyden jar having 
an equal number of square feet of tinfoil coating. There is, however, 
one material difference between the two cases. Though both are 
discharged in a time inappreciably minute to the senses, the dis- 
charge from the wire occupies a comparatively much longer interval 
than that from the coatings of the jar. 

A wire on insulating supports, in the open air, when it is uncon- 
nected with the earth, receives also a charge, but very much smaller 

1 Annales de Chimie et de Physique, troisieme serie, tome xxix. 1850. 

2 Proceedings of the Royal Institution, January 20, 1854, ' On Electric 
Induction Associated cases of current and static effects.' 


in amount, the inductive action of distant surrounding bodies exert- 
ing but little influence upon it. 

Although certain general conditions of this inductive action in 
telegraphic wires have been made the subject of experiment and are 
now well known, our knowledge in regard to this subject still 
remains very limited. 1 

The subject of inductive effects received further consideration by 
Professor Hughes who read a paper at the Society of Telegraph 
Engineers in March 1879 : ' Experimental Researches into means for 
preventing induction upon lateral wires.' Professor Hughes said 
that the induction upon lateral wires had of late years been a serious 
question upon telegraph lines the constantly increasing number of 
wires upon the same poles, added to the adoption of high speed and 
consequent sensitive apparatus, rendered the study of these effects 
of the first importance. In 1868, at the desire of the French Tele- 
graph Administration, he undertook a series of practical experiments 
with a view to finding a remedy, and he at once perceived that 
the question was of a more complicated nature than was at first 

We found then that we had to deal with the static charge of its 
own line and the dynamic induction of the lateral wires ; that the 
effects of each were very different ; and, whilst it was easy to deal 
with the comparative feeble static charge, the more powerful and 
rapid effects of dynamic induction could hardly be suppressed. 2 

Other duties prevented me at that date from following up this 
line of research. The disturbances on the lines from this cause have 
been on a constant increase from the adoption of more rapid, and 
consequently more sensitive, organs ; and in the telephone we have 
at last arrived at an organ of rapidity and sensitiveness, which not 
only reveals the constant induction, but which is the main reason 
why the telephone has not been more largely adopted upon 
telegraph lines. 3 

The conclusions reached from his experiments were that, 

if a telegraph line or a telephone line had a return wire upon the 
same poles, and absolutely equidistant from the inducing wire, we 
should have perfect protection, from the fact that the primary 
would then induce parallel currents in both wires in the same 
direction, but contrary to itself, and these parallel currents would, 
being of equal force, neutralise each other. 4 

Professor Hughes expressed the opinion that ' the cost of a 
double wire would probably prevent its use ' 8 and more attention 

1 From Report of Joint Committee, Eyre & Spottiswoode, 1861 ; quoted in 
Wheatstone's Scientific Papers, pp. 168-9. 

2 Journal of the Society of Telegraph Engineers, viii. 163. 

3 Ibid. p. 164. ' Ibid. p. 166. * Ibid. p. 167. 


was perhaps given to the method of limiting inductive effects by 
a metallic screen. During the discussion on Professor Hughes's 
paper, Sir William Preece related the results of experiments he 
had made which demonstrated that a wrapping of lead-foil or tin- 
foil considerably diminished the induction between wire and wire. 
Before Hughes demonstrated the scientific basis for the return wire 
equidistant from the inducing wire, David Brooks suggested a twisted 
double wire in cables as a preventive against inductive disturbance. 
There were apparently two types of twisted wires in use according 
to the particulars of Brooks 's cable given by members of the 
National Telephone Exchange Association. At the first meeting 
Mr. Barton said that the wires had sometimes been double, with 
one of the double wires connected to the ground. The object of 
this double wire was stated to be to avoid induction, 1 and the use 
of the term ' double ' would imply that the two wires were of equal 
gauge as in twisted pairs ; but another form was described at the 
second meeting of the Association in 1881 by Mr. Haskins of 
Milwaukee, who said : 

We have three hundred wires in six cables fifty wires to the 
cable, of the Brooks form, across the rivers. . . . Each conducting 
wire is insulated with cotton, and then a very light copper wire, in- 
sulated with cotton, is wound spirally from one end to the other on 
the outside of that. 2 

From this description it must be inferred that the conductor 
was not twisted with a wire of corresponding diameter, but remained 
straight and was surrounded by a helix consisting of a wire of 
appreciably smaller diameter than the conductor itself. This 
method appears to have been efficient in practice, for Mr. Haskins 
continues : 

I will say for those cables that the wires which run around them, 
the spiral wires which go to the ground at each end, entirely prevent 
induction between the wires. 2 

But this method is not the equivalent of the twisted conductor 
so arranged that the two limbs of the line should be equidistant 
from any source of disturbance. This feature, insisted upon by 
Hughes, is quite clearly expressed in David Brooks's patent. 

The proximity of the two telephone wires to each other is such with 
reference to neighboring wires, which might cause disturbance in 
the telephone circuit that such disturbing effect will be neutralised, 

1 National Telephone Exchange Association Report, 1880, p. 46. 

2 Ibid. 1881, p. 27. 


or in other words, the inductive effects in one wire will be neutralised 
by the inductive effects produced in the other wire of this telephone 
circuit. 1 

Brooks describes and illustrates both parallel wires and wires 
which are ' twisted round each other.' 

Bell also devoted attention to the twisted pairs, and is with 
Brooks quoted by Hughes in the paper referred to. In a note 
Hughes says that Bell had lately written him that the date of his 
patent was November 1877. In a specification 2 for a twisted 
pair cable Bell disclaims the arrangement of the wires for obviating 
inductive disturbance and the arrangement of the wires in twisted 
pairs, ' for these matters I have claimed in my application for Letters 
Patent for improvements in telephone circuits filed June 10, 1878, 
of which the present application is a division.' 

In the course of correspondence on this subject Mr. T. D. Lock- 
wood, writes me : 

I think there can be no doubt that the first suggestion of a 
metallic circuit in telephony is Professor Bell's British patent 
(which by the way has no drawing) No. 4341, November 20, 1877. 
Professor Bell, after getting this British patent, applied for a patent 
in the United States, as did also David Brooks of Philadelphia. . . . 
The U.S. Patent Office held that Bell should be put into interference 
with Brooks and was not entitled to go any further back for the 
date of his invention than the earliest date of his British patent. 
Consequently the interference was decided in favor of Brooks, who 
obtained U.S. patent No. 238,195, dated February 22, 1881. Bell, 
however, did succeed in getting from the ruins of his case a specific 
kind of metallic circuit cable patent, i.e. No. 244,426, July 19, 1881. 

The application for a British patent and the omission to apply 
at the same time in the United States, is probably explained by 
the fact that Bell was then in England. Under date of December 20 
1877, he wrote from 57 West Cromwell Road, South Kensington, 
to Preece, advising the receipt of some rheostats and condensers, 
and adding : 

I send double wire for experimental telephones. Please take 
what you wish and return the remainder as I want to make it 
into experimental cables.* 

In an article on ' Electric Communications,' published in 

1 U.S. sptcification, No. 238,195, February 22, 1881 (application filed 
March 4, 187!-'. 

a U.S. specification, No. 244,426, July 19, 1881. 

8 I am indebted to Mr. Llewellyn Preece for this letter. 


Chambers's Papers for the People in iSsy, 1 allusion is made to the 
extensive development and application of telegraphs in the United 
States. The almost limitless breadth of territory, the author says, 
necessitated a proportionate extension of ' the metallic indicator of 
intellectual supremacy,' the lines in many instances being carried 
across the country, regardless of travelled thoroughfares, over tracts 
of sand and swamp through forests and solitudes. The writer of the 
article further remarks that ' Economy and rapidity of construction 
are prime desiderata in America.' 

The telephone was following the precedent of the telegraph 
whether or not its promoters realised it. Not in solitudes, but in 
the very centre of the busiest communities, telephone exchange 
lines had their beginnings, but with them, as with the telegraph 
in the United States and elsewhere, ' Economy and rapidity of 
construction were prime desiderata.' Telephone exchange service 
was a new service. Its value had to be demonstrated. The larger 
the number of subscribers the greater its value to all. Hence 
the need of rapidity of construction. The need of economy was 
even greater, for, until the value had been demonstrated and a 
demand created, funds could not be obtained for financing the 
enterprises. These are ample reasons for the original adoption of 
single lines. Metallic circuits would have nearly doubled the cost, 
and in congested districts would have halved the numbers to whom 
service could have been afforded. 

When, therefore, cables were sought as aids to conduct the 
wires into the central offices or to facilitate the construction along 
congested routes, one of the first sources of trouble was that arising 
from overhearing. Two principal remedies were applied : the use 
of double wires in the cable, of which one was used as a conductor, 
the other earthed ; or a single conductor was wrapped with lead- 
or tinfoil. The latter type became known on both sides of the 
Atlantic as anti-induction cable and was expensive to manufacture. 

To provide a remedy in a more economical way was the pur- 
pose of W. R. Patterson's United States patent No. 253,501. 
In the cable described in this specification the conducting wires 
were single, but there was, in addition to the line wires, a central 
wire of much larger diameter than the line wires. That larger wire 
was connected to earth at both ends. The theory prompting the 
design was thus described by the patentee : 

The establishment of a current in any conductor causes an 
induced current in the opposite direction in all the other circuits 
including the low resistance circuit. It follows from Ohm's law 

1 Vol. ix. p. 27. Partly reprinted in Railways, Steamboats, and Telegraphs. 
1868, p. 27. 


that, the current being equal to the electro-motive force divided 
by the resistance, a much greater tertiary current will be induced 
from the circuit of low resistance than from any of the others. I 
find that when the resistance of this circuit is very low in comparison 
with that of any of the others say from one-tenth to one-fifteenth 
or less this tertiary current will practically neutralise the secondary 
currents in the small conductors. 1 

The theory even then was questioned, but it was found in 
practice that, when the circuit of the low resistance wire was closed, 
there was no important inductive disturbance, and when it was 
opened the ordinary effects (overhearing) were observed, 2 and the 
cable was largely used. 

The inclusion of this low resistance conductor in a cable was 
somewhat analogous to the extra wire over routes of open wires 
as mentioned by Mr. Pope, 3 and an earth connection contiguous 
to the conducting wires was early adopted for telegraph lines. One 
such plan is that of Mr. Highton, who proposed 

to place between wire and wire a direct communication with the 
earth, so that any of the electricity transmitted, as it escapes 
from the wire, may be intercepted by this communication with 
the earth, and so transmitted direct to the earth, without the 
possibility of entering an adjoining wire. 4 

Cables of the lead-pipe pattern were growing in favour and 
becoming firmly established daring the first ten years of the 
telephone's use. Whether of paraffin or hydro-carbon insulation, 
of central wire or anti-induction formation ; whether called Patter- 
son, Waring, Brooks Junior, Berthoud Borel, or other names, they 
all relied upon lead covering and material insulation. They were 
yet to undergo important changes, but the later developments in 
cables are reserved for consideration in Chapter XXII. 

1 U.S. specification, No. 253,501, January 17, 1882 (application filed 
July 14, 1881). 

2 National Telephone Exchange Report, 1882, p. 92. 

3 P. 242. 

* The Electro Magnetic Telegraph, by Laurence Turnbull, M.D. Phila- 
delphia, 1853. 



IN the preceding chapters it has been attempted to record the 
principal features in the invention and development of the telephone, 
of telephone exchanges, and accessory apparatus during the ten 
years which had elapsed since the first exchange was started, 
together with the prior work which has a relation thereto. They 
record mainly technical progress. It is proposed in the present 
chapter to consider the progress made in the telephone exchange as 
an industry a public service enterprise during the same period. 
Since it is not possible to deal with the various features either 
strictly in the order of dates or of localities, somewhat abrupt 
changes chronologically and geographically will be inevitable. 

The rudimentary central office systems connected with E. T. 
Holmes's burglar alarm system in Boston l and with Isaac Smith's 
Hartford Drug Store 2 had prepared the way for the introduction 
of specially constructed exchanges such as that which opened for 
general commercial business at New Haven, Connecticut, 3 on January 
28, 1878. 

By December 31, 1887, Boston, with the neighbouring towns 
included as branch exchanges, had 5767 subscribers, Hartford 1176, 
and New Haven 1393. 

The development in the United States was so rapid that on 
March i, 1880, 138 exchanges were in operation or about to open; 
on February 28, 1881, the number was 408, and by the end of 1886 
there were 736. 

The energy put into the business, and the promptitude with 
which available places in the United States were offered exchange 
facilities, may be observed from a statement in the first Report of 
the American Bell Telephone Company, dated March 29, 1881, 
that there were then only nine cities of more than 10,000 inhabitants 
and one of more than 15,000 without a telephone exchange. 

1 Chapter viii. p. 69. a Ibid. p. 72. 3 P. 96. 



That this covering of the ground was due to initiative and enter- 
prise and did not arise from the public demands may be observed 
from the Report for 1884, which states that as a rule all the larger 
exchanges had a steady growth, and there seemed no reason to 
doubt that that would continue for some time to come. On the 
other hand, there was a pause in building exchanges in small places, 
and some seventy-eight of those already started had been for the 
present given up, while sixty-one new ones had been established. 
The view is expressed that 

the establishment of these systems in small towns was probably 
pushed too rapidly, in view of the stagnation in general business 
which followed. Many of those now abandoned will be restored 
upon a revival of business, and others can be put into operation 
under a system which is being worked out for small exchanges 
without a central office, and which, if successful as we hope, will 
carry the telephone into a large number of towns and villages where 
it is now impossible to place them upon a paying basis. 1 

The numbers of subscribers in the United States at successive 
periods were as follows : 

February 20, 1880 . . 30,436 

December 31, 1880 . . 47,880 

1881 . . 70,525 

1882 . . 97,735 

1883 . . 123,533 

1884 . . 134,601 

1885 . . 137,570 

1886 . . 147,068 

1887 . . 158,732 

At the last-mentioned date there were 743 main and 444 branch 
exchanges, 127,902 miles of wire on poles, 9458 miles of wire on 
buildings, 8009 miles of wire underground, 363 miles of wire under 
water, and 6182 employees. The estimated number of exchange 
connections for the year 1887 was 369,203,705. 

Classified in cities of specified populations, the exchanges and 
subscribers at December 31, 1887, were : 

No. of 

No. of 

Cities with population exceeding 150,000 
,, from 50,000 to 150,000 
,, from 10,000 to 50,000 
,, under 10,000 .... 







1 American Bell Telephone Company's Report, 1884, p. 4. 



The above figures are summarised from the Company's state- 
ment of exchange statistics as at January I, 1888. The number 
of subscribers (158,712) is less by twenty than those given in the 
Annual Report (158,732). 

In Canada there were 2082 subscribers connected with the 
various exchanges at the end of the year 1880, 9614 in 1885, and 
11,600 in I886. 1 

The following represents the development in the United Kingdom 
at the nearest available dates to the end of the year 1887 : 




United Telephone Co. (London and 

district) approximately 




National Telephone Co. (Scotland, 

Midlands, and North of Ireland), 

June 30, 1887 .... 




Lancashire and Cheshire Telephonic 

Exchange Co., June 30, 1887 . 

37 88 



Western Counties and South Wales 

Telephone Co., December 31, 1887 




South of England Telephone Co., 

April 30, 1888 




Northern District Telephone Co., 

December 31,, 1887 




Telephone Co. of Ireland 


Sheffield Telephone Co. approxi- 

mately ..... 




Post Office, March 31, 1888 





The establishment of exchanges in France 2 and Belgium 3 by 
years was as follows : 

1879 Paris (September 30). 

1880 Lyon (October 15) and Marseilles (December 15). 

1881 Nantes, Le Havre, and Bordeaux. 

1882 Lille. 

1883 Reims, Roubaix, Tourcoing, Calais, Rouen, Alger, Oran, 

and St. Quentin. 

1884 Halluind, Troyes, Dunkerque, Elbeuf, Nancy, and St. 


1 American Bell Telephone Company's Reports, 1880-5-6. 

* Extracted from Montillot's Telephonic Pratique, 1893, p. 436. 

8 Extracted from La Telephonic, par Emile Pierard, 1894, p. 343. 


1885 Armentieres. 

1886 Boulogne, Cannes, Amiens, Nice, and Caen. 

1887 Fournies and Chalons. 

In Belgium at the end of 1884 there were seven exchanges 
Anvers, Bruxelles, Charleroi, Gand, Liege, Louvain, and 

There were no additions during 1885. In 

1886 Ostende Middlekerke Nieuport, La Louviere, Mons, and 
Namur ; in 

1887 Courtrai Iseghem, Malines, and Termond Alost Lock- 

eren St. Nicolas, were opened. 

Wietlisbach, the Director of Telegraphs at Berne, gives 1 the 
following particulars of the numbers of subscribers and the ap- 
proximate rates of subscription in the principal countries on the 
European Continent at the end of the year 1885 : 

Number of 

Number of 

Rates of 









8346 190-250 


(2054) 250 

France . 


7175 25O-60O 


(4054) 600 






(2326) igO 

Russia . 


528O I9O-60O 

Switzerland . 








Austria . 




















The progress made in the various countries at the nearest dates 
available to the end of 1887 is indicated in the following table 
summarised from Brault, 2 except the figures for Belgium which 
are taken from Pierard. 3 

1 Traitt de Telephonic Indnstrielle, 1888, p. 244. 

* Hisloire de la Telephonic et Exploitation des Telephones en France et 
l'tranger, par Julien Brault, 1890. 

8 La TeUphonie, par mile Pierard, 1894, p. 343. 




Number of 

Number of 


Oct. 1887 



Sweden .... 

Dec. 31, 1887 



France . . . . 1887 , 


Italy . . ... 

Oct. 1887 25 9,183 

Russia .... 

Sept. 1887 




Jan. i, 1886 



Belgium .... 

Dec. 31, 1887 


Austria .... 

Sept. 30, 1887 

13 4,200 

Holland (estimated) 

Dec. 31, 1887 ii 4,000 



22 2,677 

1 Spain .... 

Oct. 1887 8 2,200 

Portugal .... 

Dec. 1887 



Comparing the two tables it will be observed that, placed as the 
countries are in the order of numbers of subscribers, Sweden has 
become second, France third, and Italy fourth. The order of the 
other countries does not vary. 

Particulars are also given by Brault of the following : 


Number of 

Number of 

Norway .... 

Julv i, 1887 



Havana .... 

Oct. 31, 1887 373 

New Zealand . 

Jan. i, 1887 


Brazil .... 

Apl. i, 1885 




July i, 1885 


A copy of the daily ' Report of Connections ' in the city and 
suburbs of New York for December 14, 1880, came into my posses- 
sion many years ago. It was probably sent by the New York 
Company for the information of the London Company. The note 
is an autograph addition by ' H. W. P.' (presumably H. W. Pope). 
Both note and report will be of interest as indicating the position 
of the business at that date (see p. 272). 

The average calls per line in the city exchanges, it will be 
noted, were 5-31 and in the suburbs 2-60 for the day. 

Throughout the United States Exchanges the average daily 
calls per line for the respective years were as follows : 

1883 .... 4-85 

1884 . . . 5-18 

1885 . . . 5-43 

1886 .... 5-82 

1887 .... 6-37 




-._ s 



M O O t in t M too oo 



c^ O r^ O^ O\ co t 


*****" O 

o o ino ino t !n N co 


t V !-i b b co N 






-S u 


C (A 



3 -2 








s * 

co o*' t^- co too oc t in t^ 

t Mr f^ CO . CO t 




O!ntNocooococo t 


O CO 00 O t "-. 



o ^2 c_ < 

tttCOU-)incOMM CO 


Vo b b b b b 




H w 




"8 fc r" 

OO tO H M Tf" t t^O Cl 


r^ O M o t^ t co 




o. 3 

ttininininc^ot t 

in t* o* o^ o^ c> c 



O Q 



b M b b b 'M M 






' C S 

o .2 

O^OO t^OO co M O O M t^- 

O-OO O co t co CTvOO O 

vo t M in M oo inO M co 



80 in HI t G^ *n 
O o oo co co in 
t M M in c 














N t^^a^^O t 


O O M N O^ O 

t w in i o >n 





A 2 

o co o^ r*** HH \o ^^ "O *o w 

^2 ' 



B f" 

Cl M MM 



B v> 

o a 







**- *- 




1 1 




W ' t IH 












t>i M 





5 M' S *S ^"oo" Ss 5 8 ?1 



in M o N in o co 
O^ m in oo coo O 

M M 






s>. h 
Z 8 






-, Xi 






M M 

&< H 




>J 2" 

S 1 


i-i M N N co co ino r^ 
t oo o t^oo t>> t in o 

Cl M 


O f >> 

M M 




< w 



2 a 






O a, 



w '-S 


M "** 1 



^ c 



~3 H ^ 



^^ ^* 2 c s^ *^ ;? ^ 

""i X . <n . 



^ 1 a*^ ^ ^ rto-d'rt'rt M 

JS +J * C 



^^ 2 w 7^^ "o ^ o w 




on? 2 ^ ^ * rt 
C, c 2 ^ rt ii 



PQ (^ ^ M~^^ _ ^ ^_ ^^ 

_0 <3 ^ 


In the critical ten years now under consideration, telephone 
service, which was daily proving itself to be of enormous public 
benefit, was being developed with energy and resource against 
difficulties of varied kinds. Wherever official or public action was 
taken it was repressive, calling for the exercise of additional energy 
on the part of the promoters to overcome the artificial resistance 
inserted by public authorities against the advancement of a public 

Much was done, but much more might have been done if the 
public and official bodies had either been merely quiescent or had 
studied the problem more carefully. In the United States there 
were no fundamental restrictions as in Great Britain. It was 
rather on the questions of overhead wires and of rates that local 
or State authorities harassed the telephone companies. The 
New York Legislature passed an Act requiring telephone com- 
panies in New York and Brooklyn to place all their wires 
underground before November I, I885, 1 a time when no practic- 
able system had been evolved which would permit the working 
of an entire telephone exchange system through underground 

In April 1885 the State of Indiana passed a law restricting the 
price of rental of a telephone to three dollars per month. As it 
was impossible to conduct the service at that rate, the Central Union 
Company was obliged to close its exchanges in the principal cities 
and large towns, 2 the public being put to serious inconvenience 
through the act of its own government. 

A Bill to repeal the law was passed by the House of Repre- 
sentatives in the following year, but did not pass the Senate owing 
to the arrest of business from other political questions. In 1888 
it was unconditionally repealed by a large majority in both houses. 
In the meantime 3 it had ' stopped absolutely any extension of 
business, and checked that steady improvement which is constantly 
going on towards greater efficiency and convenience to the public 
in the conduct of the business.' 

Such cases as these indicate the futility of public action in 
ignorance. The Telephone Company asked that the question should 
be studied, claimed that it was worth studying : 

The telephone situation is worthy of more careful attention than 
it has received at the hands of the public. Its features are peculiar 
and the development of telephone facilities is certainly of high 
importance to our people. Although in the ten years that have 

1 American Bell Telephone Company's Report, 1884. 

2 Ibid. 1885. Ibid. 1886. 


passed since the invention became public, 156,000 miles of telephone 
wire have been built, over which 275,000,000 communications now 
pass annually, we are in reality only at the threshold of the business. 
It is possible already to talk with ease between Boston and Phila- 
delphia over our experimental wire, yet the connection of our 
principal cities and large towns by thoroughly practical telephone 
systems has, in fact, only been begun. 1 

But for the confidence and persistence of the promoters the 
telephone system, even of the United States, advanced as that was, 
would have been seriously hampered by public action. But the 
promoters were confident in the present and even more confident 
in the future : 

With the improvements that are rapidly coming into use, the 
aim must be nothing less than to provide a complete working system 
throughout the United States, which will give facilities for instant 
conversation between all points within many hundreds of miles 
of each other, such as is now possible within the limits of a single 
exchange. 1 

The arrangements with the operating companies had been made to 
this end ; the telephone companies throughout the land had been 
held under such a general plan by the force of a Government patent. 
In no other way, it was urged, could even the telephone development 
then attained have been reached so soon. And the Telephone 
Company asked the public to regard the matter as one in which 
the public were interested. There were rights of property in patents 
and in security of trade ; there were advantages to the public in 
improved means of communication. The Company asked for no 
privileges ; they only pleaded for the maintenance of ordinary 
rights : 

To anyone who will think what it would mean if instant verbal 
communication could be had by any person here with every city and 
town in New England, and as far as Philadelphia, perhaps even 
Washington and Chicago, with similar facilities in all parts of the 
country, it must be clear that this is something well worth the risk 
of money, thought, and labour, by the companies ; well worth 
encouragement by the people. The work that has been done is at 
least some guaranty of what will follow, if the protection promised 
by our patents is not interfered with. The public is ripe for the use 
of such a convenience, the inventive talent of the country is pro- 
viding new methods for making it available ; will the legislatures 
interpose to prevent all progress, through misconception of the 
problem, or because misled by interested parties ? 

1 American Bell Telephone Company's Report, 1885. 


What has already been accomplished has been under every 
discouragement that can be thrown upon a property resting on 
patent rights. But a small percentage of patented inventions prove 
valuable, and all such are usually compelled to run the gauntlet 
of law suits from pretenders, so long as their patents live. Capital, 
therefore, is timid about such investments, and nothing but the 
expectation of large temporary returns will bring it out for the 
development of a business like this. 

While it is true considerable returns have been received upon 
our property since 1883, it is equally true that for years we were 
without dividend or other means from our investment ; our patents 
have but seven more years to run, and we are still working, as always 
before, under a heavy fire and an expenditure for defence which 
nothing but a prospect of liberal returns would, for a moment, 

To undertake such a work as has just been outlined, large sums 
are required for construction. It is within the power of State 
legislatures to so alarm capital that might be attracted to such a 
business, that it would be wholly useless for a time to attempt any 
important work of the kind suggested. The attack upon rates is 
one of the most direct methods of removing all inducement to extend 
telephone facilities ; but, even if the right to absolutely regulate 
this matter were conceded, the question whether the few thousand 
persons, who now are connected with telephone exchanges, shall 
have telephone service a little cheaper than at present, is of trifling 
importance to the country compared to the completion of a tele- 
phone system adequate to the needs of the whole people. 1 

This is a somewhat long quotation, but I make no apology for 
recalling it, for it gives us a picture of the period which is of great 
value. In spite of occasional and local difficulties the Company suc- 
ceeded in accomplishing much of its object, and if reason and logic 
had been useful public aids might have accomplished more sooner. 
The telephone must triumph, even over public neglect and public 
resistance such as it received in part in the United States. But 
its difficulties there were local or ones of detail. In general the 
companies could go forward with their business in their own way 
without any restrictions from the Federal Government. 

In Great Britain the Telephone Exchange service started under 
restrictions. The first report of the United Telephone Company 
to April 30, 1881, states that an agreement had been come to with 
the Government and records that : 

There are a great number of conditions and limitations imposed, 
which tend greatly to restrict the use of the telephone for ordinary 
business purposes. From the point of view of the authorities of the 
Post Office, this may no doubt be necessary and justifiable. The 

1 American Bell Telephone Company's Report, 1885. 


Directors do not complain of it. On the contrary they have to 
acknowledge that so far as their negotiations with the Department 
have already gone, they have been met in a spirit of enlightened 
liberality, as could fairly be expected when it is remembered by 
what considerations the authorities must be hampered in their 
desire to protect the Post Office telegraph monopoly acquired at so 
great a cost. But the Board believe that experience will speedily 
prove the great superiority, for many purposes, of the telephone over 
the telegraph in facility of transmission, dispatch, and economy, 
and that it must be soon largely employed by the Post Office itself 
as a useful and indispensable auxiliary to their important business. 1 

In London, as in Boston, the great advantages of the telephone 
were recognised by those who had undertaken to exploit it. They 
knew that its use must be restricted by the conditions and limita- 
tions imposed. The principal limitation was the one of distance, and 
the object of the limitation was to conserve the local telegraph 
revenue. This condition remained in force until 1884. Its bearing 
may be observed from the report of the Telephone Company of 
Ireland, whose chairman (Mr. Dwyer Gray, M.P.) rendered important 
service in assisting to obtain its removal. In the report of that 
company to June 30, 1884, is to be found the following statement : 

That the telephone supplies a want not met by the telegraph 
namely that of obtaining an instantaneous reply is demonstrated 
by the fact that the gross rental of the company from the Dublin 
district alone is now already upwards of 8000 per annum over 
an area with a radius of five miles from the central Post Office, 
whilst the amount received by the Post Office Department from 
telegrams despatched from an area with a radius of twelve miles 
from the Post Office to addresses within the same area, as stated 
by the Postmaster General in the House of Commons on June 9, 
1884, was only 1390 per ann. 2 

The Irish Company is the only one recording this feature of 
comparison, but it cannot be regarded as typical. In other localities 
there was probably a greater local telegraph business, but the 
restriction must have had a very retarding effect on the adoption 
of the telephone exchange service by the public. In the new license 
the limitation of distance was withdrawn. The effect of its removal 
and the expectations of the directors thereon are thus recorded in 
the same report : 

The new license will permit the company to open public call- 
offices, and also to give intercommunication between towns. The 
directors anticipate that these privileges and conveniences will 

1 United Telephone Company (London) Report, 1881. 
* Telephone Company of Ireland Report, 1884. 


tend greatly to facilitate business and be of considerable advantage 
to the company. 1 

The Governmental restrictions were not the only ones the 
British companies had to strive against in their efforts to introduce 
a public service. Local authorities sometimes helped to retard the 
work. The report of the Northern District Company for the year 
ending December 31, 1882, contained the following paragraph : 

The directors feel that this increase [in number of calls] would 
have been very much larger but for the difficulty and delay caused 
by the opposition of some members of the Council who are strongly 
represented on the Highways Committee. This difficulty would 
be no doubt very much lessened if it did not altogether disappear 
if the business men of the town would impress upon their repre- 
sentatives in the Council the necessity of giving every facility for 
cheap and improved telephonic communication. The action of 
the Corporation prevents the company connecting quickly those 
who are desirous of joining the exchange, which is not only injurious 
and inconvenient to the member who has to wait, but to all the 
other exchange subscribers. 2 

The improvement effected by Mr. Fawcett's removal of the 
four-mile limit 3 is reflected in the report of the Lancashire and 
Cheshire Company for the year ending June 30, 1885. This stated 
that the average daily calls made by subscribers still maintained 
its steady increase, which ' is the best evidence that the telephone 
service is giving satisfaction.' In several of the smaller exchanges 
there had recently been a marked improvement both in the number 
of inquiries for telephones and in the average of calls made. 
' No doubt the new trunk-wire and call-office services will have 
largely aided in awakening in these outside towns this increased 

A lack of knowledge on the part of the public of the value of 
the exchange system is indicated by a suggestion made by a share- 
holder at the Annual Meeting of the South of England Company 
(1888) that ' the directors might by means of lectures, at the cost 
of a few pounds make the uses of the telephone more generally 

But in spite of all the drawbacks placed in its way, telephone 
exchange service was progressing. Public appreciation was growing, 
subscribers were increasing. To the larger class of business people 
telephonic communication was becoming more of a necessity and 
less of a luxury. The use of the service was still within a very 

1 Telephone Company of Ireland Report, 1884. 

2 Northern District Telephone Company Report, 1882. 

3 Chapter xxxii. p. 504. 


restricted field. But the field was ever extending its boundaries. 
Demonstrations of its usefulness to the few were becoming apparent 
to the many, and subscribers were increasing accordingly. Improve- 
ments in the technical features had followed the demands of the 
business as experience forced those demands upon attention. Co- 
operative efforts towards improvements were in some measure 
available through the exchange of experiences at the annual 
conferences or conventions of the National Telephone Exchange 
Association, but improvements mainly resulted from the inventive 
ingenuity of telephone engineers or the individual discoveries of 
scientific men in America or Europe, and of the efforts on the 
part of manufacturers to keep abreast of the requirements of their 

From the first those responsible for the initiation of telephone 
exchange business saw that it was to be of great public benefit, 
and endeavoured to impress the public with the same idea. But 
the service was new, and they equally well realised that they had 
to learn their business in the school of experience. They placed no 
limit ultimately on distance or development, but only asked that 
they should not be called upon to carry out works of construction 
that were physically impossible in the then state of knowledge. 
But the school of experience was teaching much. 

Advocates of systems were being subjected to the practical test 
of results as well as the criticisms of their colleagues at the annual 
conventions previously referred to. These conventions were notice- 
able for the full and free information mutually exchanged. But at 
the end of the period to which reference is now made it had become 
apparent to the management of the American Bell Telephone 
Company that the future growth would be such that provision could 
only be made for it satisfactorily by a standardised system of plant 
and apparatus, along lines which should commend themselves to 
the most capable and experienced minds in the business. Accord- 
ingly, during the year 1887 the company arranged for the meeting 
of conferences of experts on cables and switchboards, thus inaugur- 
ating a method of organised co-operation in the study of problems 
and their solution, which had important results in the development 
of the industry. 


(THE CABLE CONFERENCES OF 1887, 1889, AND 1891) 

THE report of the 1887 Cable Conference l (' printed ' but ' not 
published ') covers 152 pages. 

The purpose of the conference and the state of information at 
the time are admirably set out in Mr. E. J. Hall's opening words 
from the chair, as follows : 

We meet here to-day for the purpose of discussing in an entirely 
informal manner the subject of telephone cables. We all realise 
how great the need is for more information on this subject, and 
how few sources of information are open in electrical literature or 
in the current published records of telephone work. 

Looking at the matter from different standpoints, we wish to 
add together the results of our observations and experience, to 
discuss the theories which explain observed phenomena, and to 
illustrate by practical tests the application of the formulas which 
we hope to have presented, for our practical use in the future. 

It is, perhaps, too much to expect that we can now reach the 
desired result of formulating definite rules covering all the details 
of specifications for telephone cables for any stated use. The art 
of telephony is probably too new to have yet reached any such 

1 The members were : 

Edward J. Hall, General Manager, American Telephone and Telegraph 
Company (in the chair). 

E. M. Barton, President, Western Electric Company. 

W. W. Jacques, Electrician, American Bell Telephone Company. 

W. R. Patterson, Electrician, Western Electric Company. 

John A. Barrett, Electrician, American Telephone and Telegraph Company. 

A. S. Hibbard, General Superintendent, American Telephone and Tele- 
graph Company. 

\V. D. Sargent, General Manager, New York and New Jersey Telephone 

W. H. Eckert, General Superintendent, Metropolitan Telephone and 
Telegraph Company. 

Joseph P. Davis, Consulting Engineer, American Bell Telephone Company. 

H. B. Thayer, Manager of the Western Electric Company in New York. 



state of exact knowledge. We may, however, agree on some pro- 
positions that will so far help in our work as to keep us progressing 
in the right direction and enable us to avoid some of the costly 
experiments involved in testing, as we have all been obliged to, 
untried hypotheses on the large scale of practical work. 

If the results of our conference show that we have not yet 
reached any broad basis of agreement, they will serve to emphasise 
the necessity of continued investigation and study, and we will have 
in the report of this meeting a large amount of valuable data 
preserved in a permanent form available for future reference. To 
this we may hope to add from time to time the results of our own 
broader experience, and the mass of information which is daily 
being accumulated by the many able observers who are working in 
all parts of the telephone field. 

During the discussion on Thompson's paper on ' Telephonic 
Investigations ' at the Society of Telegraph Engineers (I.E.E.) 
on February 10, 1887, Preece propounded a law to determine the 
limiting distance of speech : 

We take the total capacity of a circuit (K), which is found by 
multiplying the length of line by its capacity per unit length, and 
the total resistance (R), which is found by multiplying the length 
of line by its resistance per unit ; and the result is that we get a 
law determining the distance to which we can speak that is simply 
expressed by the product K by R. 1 

Professor Thompson in his reply objected to the comparison of a 
telephone line to a submarine cable in which capacity and resistance 
were the only things to be taken into account. He, on the other 
hand, held that in a telephone line the important things were the 
mutual induction, the self-induction, and the resistance ; the 
capacity being negligibly small. 2 Professor Ayrton during the 
same discussion contributed valuable remarks on the factors to be 
considered on telephone lines and cables, of which the details are 
available for ready reference and do not therefore need quotation. 
It suffices to call attention to the fact that early in 1887 scientists 
were considering what had previously received but little attention 
the conditions which controlled the transmission of the electrical 
equivalent of vocal vibrations in the line and the improvements 
to be sought. 

In was in September 1887 that the conference of specialists 
sat in New York to inquire into the subject of cables and to advise 
the American companies upon the most suitable types to adopt. 
The first paper was by Mr. Jacques, who appears to have adopted 

1 Journal of the Society*of Telegraph Engineers, xvi. 84. * Ibid. p. 135. 


the views of Preece in part. In the course of his paper he stated 
that : 

No matter what may be the distance between two points 
good business conversation may be carried on between them, 
provided they be connected by a pole line or cable or both, the 
product of whose total resistance by its total capacity is less 

than 4500 if transmitters of the Hgs type be used. 

He added, ' This rule is purely the result of experiment.' 

Mr. Jacques' rule was subjected to some criticism, that by Mr. 
Patterson being selected for quotation as being the most illuminating 
and borne out by subsequent experience. Mr. Patterson said : 

I have been inclined to doubt the applicability of any general 
formula as simple as resistance multiplied by capacity as giving 
the correct working value of telephone circuits. First, the doctors 

Mr. Preece says that the constant varies for the different 
materials of the circuit, and gives for an overhead line of copper 
a value of i5,-ooo ; for copper cables, 12,000 ; and for overhead 
iron, 10,000. 

Dr. Wietlisbach gives as the result of his experiments and 
theorising, that if CR equals 100, the transmission is excellent ; 
if 300, it is good ; if 1000, it is possible ; if 2500, it is impossible ; 
making no variation for different transmitters. 

Dr. Jacques says that for the Blake transmitter the constant 
is 2000, and for the Runnings, 4500. I do not see how the transmitter 
can be ignored as it is by Preece and Wietlisbach, as we have every 
day practical demonstration of their difference. I am inclined to 
believe that capacity, or the factors which affect capacity, have 
more than a proportional share in producing retardation. If the 
wires are near together, not only is the capacity higher for the same 
material, but the dynamic or kinetic effect is increased and has a 
retarding effect on the current. If we connect a number of con- 
ductors in a cable in multiple arc the resistance is diminished and 
the capacity as measured is not increased in a corresponding ratio, 
but the more such conductors we get in the circuit the worse it 

Mr. Patterson, while criticising the laws sought to be deduced, 
assumed that all were agreed ' that resistance and capacity are 
the principal obstacles to telephone working, and they should be 
as far as practicable removed.' Of the three ways of destroying 
the efficiency of telephonic transmission he had the idea ' that 
resistance attacks principally the loudness, and capacity and other 
things, which depend upon the same factors as capacity, attack the 
clearness.' Approaching the subject from the point of view which 


should be ever present to an engineer how to obtain specified 
results for the lowest expenditure Mr. Patterson continued : 

In looking at the problem to see where we can get rid of the 
most resistance and capacity, we see that circuits in which we 
approach the limits are generally made up of long pole lines and 
shorter cable lines, say 100 miles of pole line and 10 or 15 miles of 
cable. The great bulk of capacity is in the cable of resistance in 
the pole line. To reduce the resistance of the cable implies that 
we must increase the thickness of insulation to keep a corresponding 
capacity, and consequently the size of the cable and conduits must 
be increased. If we reduce the resistance of the pole line the 
increase in its capacity is hardly appreciable, and there is no cost 
except the additional weight of copper. Then the short lines 
around the exchange alone will work any way. They are many 
more in number than the long pole line. The money to be expended 
in increasing the size of all these wires and in correspondingly 
increasing the insulation would accomplish much more if applied 
to increasing the size of the comparatively fewer overhead wires. 
Let the money which is to be expended upon cables go principally 
to keeping the capacity low and the wires well apart. 

The economics of the situation were here well stated and 

On the exteriors of cables evidence was given by Mr. Sargent 
as follows : 

The first lead cables that were laid were manufactured by the 
Western Electric and Brooks of Philadelphia. The former have 
now been down almost three years, and the lead pipe shows scarcely 
any signs of decay. There is on the surface a hard white crust, 
but it is very thin and does not seem to be increasing with the 
lapse of time. The Brooks cable at the end of eight months was 
found to be badly corroded, a crust of what appeared to be carbonate 
of lead forming on the outside in that time one-sixteenth of an inch 
thick, cracking and peeling off whenever the cable was handled. Six 
hundred feet of this cable was drawn out on account of defective 
insulation and replaced by another piece. This also has corroded 
badly. Otherwise the whole length of about 2400 feet is still in 
service, but the corrosion appears to be going on with every prospect 
that it will continue until the lead is destroyed. As far as my 
information goes the only difference in the lead of these two cables 
is that the Brooks claims to be pure lead, while the Western Electric 
has a small percentage of tin mixed with it for the purpose of 
hardening it. 

Based upon the results of experience, it has continued to be the 
practice of telephone companies in the United States to use the 
alloy of tin in cable sheaths, but the practice is not universal in 


Europe, the British Post Office and many continental engineers 
being quite satisfied with pure lead. The durability of lead pipes 
is amply demonstrated by the continued existence of Roman work 
of this character, of which that in the city of Bath may be taken 
as an example. It should, however, be borne in mind that the 
modern ' pure lead ' of commerce has all the more valuable in- 
gredients extracted from it, which was not the case with the material 
used by the Roman artificers. 1 The introduction of the tin alloy 
serves to restore the lead more nearly to the condition of the metal 
of whose durability we have concrete evidence. As, however, tele- 
phone cables can hardly be laid with a view to use so far in the 
future as Roman work is in the past, there is still room for argument 
as to the money value of the alloy. But in respect to tensile 
strength it has been satisfactorily determined that the alloy is 
superior, so that to obtain equal results a thicker sheath of pure 
lead must be used than would be required where tin is added. 

As the result of practical tests and theoretical calculations the 
conference recommended the use of a No. 18 Brown and Sharp 
gauge (-040 inch) conductor with insulation to bring the diameter up 
to ^ inch, which it was calculated would enable conversation to be 
carried on through twenty-five miles of cable. For single circuits 
frequent transposition of the wires was recommended for the reduc- 
tion of cross-talk. The efficacy of this method, it may be noted, was 
observed by the officials testing the Brooks cable in London in i88i. 2 
Whilst considering this expedient it was further recommended that 
any companies contemplating the introduction of metallic circuits 
in their system within the next two or three years should at once 
use cables with twisted pairs, leaving one limb of the pair idle in the 
intervening period. 

The specification subsequently prepared to conform to the 
conclusion of the members of the conference is given in full as 
follows : 


Specifications for telephone cables prepared on the basis of the con- 
clusions reached by the Cable Conference, September 1887. 

Size of Conductors. Each conductor shall be of No. 18 Brown & 
Sharp gauge, best quality Lake copper. 

Thickness of Insulation. Each conductor shall be insulated to 
125 of an inch, with not less than two wrappings of cotton so put 
on that when the conductors are laid up in a cable there shall be 
no substantial compression of their coverings. 

1 Mr. A. P. Trotter referred to this point in the course of a discussion at 
the Institution of Electrical Engineers. Journal, xviii. 365. 

2 Chapter xx. p. 254. 


Twisted in Pairs. The conductors shall be twisted in pairs, 
the twists being regular and uniform, not less than 2f inches, nor 
more than 3^ inches in length. 

Reversed Layers. The cables shall be laid up in reversed layers, 
each layer being served with one covering of cotton ; the pitch 
of each layer shall be as long as is consistent with necessary 

The cables shall be so constructed that there shall be no sub- 
stantial compression between the layers or between the outside 
layers and the lead pipe. 

Outside Coverings. The core shall be enclosed in a composition 
lead and tin pipe (for underground cables 97 per cent, lead and 
3 per cent, tin) of uniform thickness and free from holes or other 
defects, weighing not less than : 

3| Ib. per foot for if inch pipe. 

3 i* ,. 
2f ,, if 

2| Ij 


Dimensions are for inside diameter of pipe. 

Insulating Material. The spaces in the core and between the 
core and pipe shall be filled with insulating material which shall 
give an electrostatic capacity not exceeding -20 of a micro-farad per 
mile and an insulation of not less than 100 megohms per mile when 
laid and spliced and connected with terminals ready for use ; 
and such capacity shall not increase nor such insulation decrease 
beyond the limits above specified for one year after the cable has 
been laid and spliced, except from mechanical injury. 

Pated March 27, 1888.) 

The 1889 Cable Conference 1 Report is a volume of 125 pages. 
Like its predecessor, it was printed but not published. 

1 The representatives were : 

Edward J. Hall, Jr., Vice-President and General Manager, American 
Telephone and Telegraph Company. 

John A. Barrett, Electrician, American Telephone and Telegraph Company. 

J. E. Crandall, Electrician, Chesapeake and Potomac Telephone Company. 

William H. Eckert, General Manager, Metropolitan Telephone and 
Telegraph Company. 

C. H. Wilson, Superintendent, Chicago Telephone Company. 

George A. Hamilton, Electrician, Western Electric Company. 

Hammond V. Hayes, Electrician, American Bell Telephone Company. 

F. A. Pickernell, Superintendent of Equipment, American Telephone and 
Telegraph Company. 

1. H. Farnham, Electrician, New England Telephone and Telegraph 

W. R. Patterson, Electrician, Western Electric Company. 


In brief introductory remarks the Chairman (Mr. E. J. Hall) 
said : 

In September 1887 we had a conference here on the subject of 
telephone cables, at which we reached certain general conclusions ; 
the object of this meeting is to revise those conclusions in the light 
of the experience of the last two years, and determine whether the 
specifications, which were prepared on the basis of our discussions 
then, should be modified in any respect, and, if so, in what way. 

The first paper read was by Mr. Barrett, who remarked that : 

The cheapness of lead pipe cable, added to some very important 
electrical advantages which it possesses, makes it beyond question 
the best known cable for general telephone purposes. . . . 

The cheapness of the lead-pipe cable does not require discussion 
at this point, but low inductive capacity needs to be earnestly 
dwelt upon, since this is the chief burden which subterranean 
telephoning staggers under. 

This burden is becoming more and more prominent as the lengths 
of the underground cables are increased and as greater distances 
overland are brought into connection with the city cable systems. 

It is the point of chief importance to have this matter of low 
static capacity keenly and thoroughly appreciated early in the 
development of the underground cable system, so that the steps 
taken in this direction may not be to future disadvantage. . . . 

Since the adoption of this standard by the conference of Septem- 
ber 1887, 1 have, in connection with Mr. W. D. Sargent, of Brooklyn, 
been engaged in an effort to reduce the specific inductive capacity 
of the wrapping used upon the conductors so as to secure a con- 
siderably lower limit for static capacity while still using the same 
dimensions for the cable. 

We have had an almost unlooked-for success in this direction 
in the employment of manilla paper in the place of cotton as the 
wrapping for the conductors. 

Difficulty in getting our paper covering manufactured into core 
for our use has prevented our putting any large amount of this 
cable into service, or to the exhaustive tests which would enable 
us to determine as closely as we desire the difference in specific 
inductive capacity between paper and cotton ; still we have several 
lengths of cable of 500 feet each upon which we have at various times 
made tests in connection with Mr. F. A. Pickernell, of the American 
Telephone and Telegraph Company, Mr. Geo A. Hamilton, now 
of the Western Electric Company, and Mr. Chas. Matchett, of the 
New York and New Jersey Telephone Company. 

A. S. Hibbard, General Superintendent, American Telephone and Telegraph 

Joseph P. Davis, Consulting Engineer, American Bell Telephone Company. 

J. C. Reilly, General Superintendent, New York and New Jersey Telephone 


As a general deduction of these tests; without going into a detail 
of the figures, I have no hesitation in saying that, other things being 
equal, the substitution of paper for cotton in the manner in which 
we have employed it will give a reduction of 30 per cent, in static 

Mr. Hamilton has a schedule of his tests which he may be willing 
to submit to consideration with his remarks upon this subject. 

In respect to the mechanical properties of paper for cable pur- 
poses and the method of manufacture which has met our require- 
ments most satisfactorily, I will say that such paper as we have used 
is much less hygroscopic than cotton, so is easier to prepare in 
expelling the latent moisture and easier to maintain in high insula- 
tion through subsequent exposure. It can be laid upon the con- 
ductors very dense and hard and in smooth round form and with the 
requisite degree of flexibility. It is slightly lighter per cable foot 
than cotton. It retains its shape well in cabling and is convenient 
to handle in making splices and connections. Its cost per cable 
foot is approximately that of cotton. It is a little cheaper per 
pound as material and a little more expensive in the labor of 

The cable from which we have secured our good results has 
been made with core furnished us by the Norwich Insulated Wire 
Company, now of this city. We have tried various methods of 
applying the paper to the conductors, but the product of this 
company alone has had the requisite hardness, smoothness, and 
flexibility to give us to the full degree the good results we have 
sought for. 

In respect to the standard dimensions for cable, two inches in 
external diameter is the size determined upon in view of the mechani- 
cal conditions of the underground problem. There is a constant 
temptation and tendency to take advantage of every diminution 
of specific inductive capacity of material to the end of increasing 
the number of conductors placed in the given external dimensions 
and so to diminish the cost of the cable per conductor. 

Thus in the standard formerly fixed, the conductor being placed 
at No. 18 B. and S. and a maximum limit of -20 mf . per mile named, 
as soon as it was discovered that a few more than the prescribed 
number (fifty-one or fifty-two pairs of conductors) could be squeezed 
into the space, this was done instead of using the advantage to 
the reduction in static capacity to a possible -18 mf. or -17 mf. 
per mile. 

It seems to me that if this matter of the relations of static 
capacity in cables to the future of the telephone system were 
appreciated to its full extent, then every gain in reducing the specific 
capacity of insulating media would be received with satisfaction 
for its own sake alone and not for the sake of a possible cheapening 
in the present cost of the cable. And I believe that this being 
agreed to namely, that the telephone companies can afford to 
employ cables in which fifty-one or fifty-two pairs go to the two 


inches then this standard should be adhered to and the chief 
recommendations to a cable be laid upon its low static capacity. 
The standard of fifty-two pairs to a 2-inch cable was substantially 
agreed upon as the maximum limit of cost which local telephone 
companies could afford for the purposes of their local service. The 
experience of the long-distance telephone service indicates the 
necessity of a more liberal provision for long trunk lines. A single 
instance will illustrate this point, and will indicate pointedly how 
truly the static capacity of cables is a limiting function of telephonic 

The line from New York to Buffalo is approximately 475 miles, 
of which about five miles consists of the Conference Standard Cable. 
Tests indicate that this five miles of cable constitutes about one-fourth 
of the whole telephonic distance to Buffalo. The practical effect 
of this is shown in the fact that a good Blake transmitter over 
470 miles of the pole line gives excellent transmission. The same 
instrument through the line with five miles of cable added is almost 
entirely inoperative. 

In this same case of pole line alone the long-distance transmitter 
has a good clear margin of safety with which to meet the adverse 
conditions of average service, while with the addition of this five 
miles of cable the dependence of the same service upon favourable 
conditions at all points is seriously enhanced. 

On this account I should recommend for long-distance trunk 
line service a special cable standard. My suggestion for this would 
be lead pipe cable, No. 16 B. and S. conductor, paper wrapping 
and twenty pairs to the 2 inches. 

This cable would probably give about '05 mf. per mile. If 
upon trial it is found that twenty pairs to 2 inches would give more 
than -05 mf. per mile, I should reduce the number of pairs rather 
than to run over that limit of capacity. 

At a later period of the sitting Mr. Barrett said : 

In the matter of the paper material which I have introduced our 
opportunities for testing and getting results which are as full, and 
reliable as we would like, have been very meagre, and I should like 
to have that matter left more or less open, to be determined by some 
more tests that are in process of being applied. We are having 
other lengths, to a considerable extent, of paper cable now made, 
and within a month or two we shall have some very definite results 
on that point. 

Mr. Hibbard suggested that the wording of the specification 
should be so altered as to include paper or anything else which 
might be found to be better than cotton. 

The prevailing type of cable hitherto was one in which a covering 
of cotton served as the medium of separation between one wire 
and another. Mr. Barrett and Mr. Sargent experimented with 


paper with the expectation that a lower capacity would result. 
These experiments were limited to the substitution of paper for 
cotton as a separator. For insulation, paraffin wax, resin oil or 
other compound was still relied upon, and the tube was filled so 
far as possible with the insulating material. The experiments 
with the paper cable, though ' very meagre ' as Mr. Barrett said, 
bore out the expectations of the designers in showing a lower 
capacity than the prevailing type. 

Following the remarks last quoted, Mr. Patterson said : 

In regard to the use of paper, I tried some experiments some 
time ago and at the time came to the conclusion that the lower 
capacity was due to the fact that more air was contained in the 
pores of the paper and that the filling. material did not so readily 
penetrate, and that just so much as the capacity was lowered its 
power of absorbing water was increased. This was rather a hasty 
conclusion, but it was enough at the time to set me off into another 
line of work which, of course, appears rather risky and which under 
careful consideration I think results in a very good cable : that is, 
to leave the greater part of the length of the insulation dry without 
any filling material at all ; when the cable is brought into one of 
these ducts, especially when there is a large number of ducts 
together in one body and properly protected, there is almost an 
infinitesimal chance of mechanical injury in the length of the cable. 
About the only trouble will come in our manholes where workmen 
are engaged in drawing in other cable and splicing, and it may be 
on account of explosions in manholes. I tried some experiments in 
making cables in which the core was introduced into the pipe 
perfectly dry; then a pressure of gas was put on, say 90 Ib. 
to the inch, and after the air was compressed in the pipe, the 
filling introduced from both ends simultaneously at an increase 
of pressure of 10 to 15 Ib., so that the filling would penetrate 
for perhaps 25 feet from each end, while the body of the cable 
was left dry and without any filling. The experiments on this 
have not been very extensive, but have been such that I think 
the static capacity of cables made in that way with cotton 
would come down inside of -15 of a micro-farad per mile, and 
possibly lower than that if jute was used ; and, if paper is of 
really lower specific inductive capacity, there would be a still 
greater advantage in the use of paper. In the case of any mechan- 
ical injury to the cable between manholes, that length of cable 
has got to be pulled out and another put in its place. Now, if the 
core is dry and loose, the core can be pulled out of the pipe 
and used elsewhere ; whilst if it is enclosed in the pipe tightly it 
has got to be spliced, and a sleeve splice between two lengths of 
cable would not be very easily pulled into the ducts. Some of 
this dry cable was made last year for use in New York where we 
crossed the route of the steam pipes. 


These observations occur on page 61 of the report. The com- 
mittee continued its deliberations to the extent of another sixty- 
four pages of print without any further reference to the omission 
of filling material, and finally compiled the specification abstracted 
below, which, following that of the 1887 conference, provided that 
' the spaces in the core and the pipe shall be filled with insulating 
material/ though the electrostatic capacity specified was reduced 
from *2O to not exceeding 'i8 of a micro-farad per mile. 

The attitude of this committee of practical men in thus ignoring 
a suggestion which lay at the root of the matter is to be explained 
by the fact that confidence in the protecting envelope was not yet 
attained. Mr. Patterson put the suggestion forward tentatively, 
considered that it appeared ' rather risky/ and yet indicated truly 
that -danger from damage in the interior of the conduit was remote. 
Where danger existed was at the manholes, and here it was proposed 
that ' filling ' should be provided. The dry cable had only been 
used in practice for a special purpose not with a view to the 
reduction of capacity, but because the temperature in its neighbour- 
hood would have melted an insulating material. 

The need of a low capacity was thoroughly recognised, but the 
security of an insulating material was tenaciously held to by men 
who had ever present to their minds the serious consequences which 
must ensue from even a microscopic defect in the lead covering. 
The service of a large number of circuits would be interrupted, and 
continuity of service was already recognised as an essential. 

The 1889 specification followed generally that of the 1887 
conference given on page 283. The size of conductors was retained 
at No. 18 B. and S. gauge ; but the copper was to be ' 98 per 
cent, pure with a resistance per mile not greater than 35 ohms at 
60 degrees F. after the cable is laid and connected to terminals/ 

The thickness of insulation, the twisted pairs, the reversed 
layers, and the outside coverings were unaltered. 

The capacity was reduced to -18, ' each wire being tested against 
all the others grounded.' 

There are the following additions : 

Covering for Pipe. 

1. Th6 pipe to be thoroughly coated with asphalt. 

2. A protecting jacket to be laid on outside ; this jacket to 
consist of at least two wrappings of tape put on in reversed layers 
and thoroughly impregnated with asphalt. 

Referring to the clause providing for wrapping the conductors 
with cotton, the following resolution was passed : 

It is the sense of this conference that it is very important to 
follow up the experiments in the use of paper or other materials 
as a substitute for cotton, and that it is understood that it is desir- 


able to authorise such substitution whenever the manufacturers will 
guarantee a substantial reduction in static capacity therefrom. 

Two years later (1891) further conferences were held. The 
official designation was altered to that of the Cable Committee. 1 
The reason for the change was indicated by the Chairman 
(Mr. Hall) at the first meeting as follows : 

In view of the frequent development of new ideas in the con- 
struction of cables, Mr. Hudson has thought best to have a standing 
committee take the place of the irregular meetings which have been 
held heretofore, and this meeting this morning is intended to be 
the first meeting of that committee. 

The committee proceeded to discuss the prior specification- and 
modifications suggested therein. The discussion on the sixth 
clause indicates so clearly the starting-point of confidence in the 
envelope and the general attitude of experts previously referred 
to, that it is considered desirable to give it in extenso : 

Mr. HALL : We will take up No. 6, Insulating Material. 

6. Insulating Material. 'The spaces in the core and between 
the core and pipe shall be filled with insulating material which shall 
give an electro-static capacity not exceeding -18 of a micro-farad 
per mile and an insulation of not less than 100 megohms per mile 
at 60 degrees F. when laid and spliced and connected with terminals 
ready for use, each wire being tested against all the others grounded ; 
and such capacity shall not increase, nor such insulation decrease, 
beyond the limits above specified for one year after the cable has 
been laid and spliced, except from mechanical injury.' 

That brings up the question of what the requirement for static 
capacity should be, and I would suggest that we talk about that 
first, as this clause is a pretty long one and brings in three or four 
different subjects. The present specification has been objected to 
on the ground that it required that the space in the core and between 
the core and the pipe shall be filled with something. I understand 
Mr. Patterson has tried to convince some people that the filling 
might be dry air or anything else, but that they thought they ought 
to have some tangible substance for their money. I should think 
that you could maintain a successful law-suit on your interpretation 
of the specification. We all agree that air is an insulating material 

1 At the April meeting the members present were : 

Edward J. Hall, Jr., Vice-President and General Manager, American 
Telephone and Telegraph Company. 

George A. Hamilton, Electrician, Western Electric Company. 

Hammond V. Hayes, Superintendent Mechanical Department, American 
Bell Telephone Company. 

F. A. Pickernell, Engineer, American Telephone and Telegraph Company. 

W. R. Patterson, Superintendent, Western Electric Company. 

J. J. Carty, Electrician, Metropolitan Telephone and Telegraph Company. 


and you could probably prove that it was filled with it where it was 
not filled with something else. 

Mr. HAYES : Of course that could be left out entirely and be left 
entirely optional. I should think nothing need be said about it. 

Mr. HALL : I should think it would not be worth while to say 
anything about that now, because we have left the material, which 
is to constitute the core, practically to the manufacturer. 

Mr. HAYES : So it seems to me. 

Mr. HALL : So it perhaps might take this shape. I will read 
a suggestion that Mr. Patterson has made. 

' Insulation. Insulation resistance shall not be less than (blank) 
megohms per mile after cables are laid and spliced, or not less 
than (blank) megohms per mile measured on reel. Insulation 
measurements to be made on one wire against all the others and 
the pipe. 

' Capacity. Electro-static capacity shall not be greater than 
(blank) micro-farads per mile for dry core, and not greater than 
(blank) micro-farads per mile for filled cable. This capacity to be that 
of one wire against all the others and the pipe. The capacity of one 
wire against its mate shall be not greater than (blank) and (blank) 
micro-farads per mile respectively for dry and filled cable.' 

You separate the two things, insulation and capacity, in your 
specification. I should think that was I was going to say that 
I thought that was a wise arrangement, but I don't know that it is. 
Really this old specification applies, as it is commonly supposed, to 
a filled cable. Now you have the question before you whether you 
want to have a filled cable or a dry cable or a partially filled cable, 
and, until you have discussed that, I don't see how you can name 
any figure for capacity. 

Mr. PATTERSON : I think all the saving that has been made in 
capacity since the 1887 conference has been caused by bringing 
in air and leaving out the other filling material. It is gradually 
worked down now so that everybody is willing to consider a cable 
that has no rilling except air, and the capacity can consequently range 
all the way between something like '08 and something like -19 or 
"20, if it is thoroughly and honestly filled and saturated. 

Mr. HAYES : Have any measurements been made with dry core 
or partially dry core cables that have been down a year or more ? 
They have been down in New York here I think a year or more. 

I Mr. PICKERNELL i We had a test made of a partially dry core 
cable that is, the ends of the section were filled back 15 or 20 feet ; 
a cable of five miles long that had been laid one year. This cable 
was in substantially the same condition as it was when laid. The 
insulation at that time was something like 1000 megohms per mile. 
The capacity was the same as it was when made. 

Mr. HAYES : I have heard it suggested that it would be impossible 
to make a lead pipe so dense that moisture would not get through 
it in time, and that there must be a deterioration. Of course water 
is fatal to the life of a dry core cable. 

u 2 


Mr. HALL : Isn't it fatal to the life of any cable ? If water can 
get through the pipe it will get through the core anyway. 

Mr. CARTY : I think our submarine cable experience would 
give us as severe a case of that as we could get. 

Mr. PICKERNELL : We have used Patterson's submarine cable 
three or four years and never have had but one case of trouble with 
it, and that was caused by a pickaxe at the shore end. The 
insulation is as good as when first laid. 

Mr. HAYES : If that has been the result of experience, then why 
shouldn't we use dry core cables ? 

Mr. PICKERNELL : I don't see any place where we could use filled 
cables to advantage, not when they are drawn into the conduits. 
In case of injury between manholes a length has got to come out, 
and there is no more risk in dry core cables than in filled core. 

Mr. PATTERSON : I have been about twelve years combating 
the idea that there are pores in lead big enough for moisture to get 
through. I never saw water inside of a pipe that did not get through 
a hole, not through microscopic pores. The original inventor of the 
dry core cable is Mr. Brooks. He patented cables with dry air, 
keeping air under pressure. He laid a cable in that way and found 
moisture inside, and came to the conclusion that it sweat on the 
inside as he had seen pipes sweat on the outside, and he would not 
have anything more to do with air as an insulator. 

Mr. CARTY : It is my belief that there never has been a cable 
failure not caused by mechanical or chemical injury to the lead 

Mr. PATTERSON : Or some original imperfection. 

Mr. CARTY : Mr. Pickernell has had some experience with 
some faults in cables in Boston, which seem to show that the filling 
is not of such great importance. Was it in Boston ? 

Mr. PICKERNELL : I don't know that it showed that. We had 
some fault in a Patterson cable ; there being a pinhole in a wiped 
joint where it was entirely submerged in water . . . about one- 
half of the conductors were affected ; the other half were in good 
condition. That was a filled cable. . . . the moisture did not 
penetrate very rapidly even when water was in the pipe. . . . We 
had rather an interesting experience with a dry core cable in 
Philadelphia. That cable ran at one point through an open sewer ; 
men at work punched two or three holes through it with a pickaxe ; 
. . . one-half of the conductors had an insulation at that time of 
600 megohms to the mile, and the other half had an insulation of 
less than 10,000 ohms to the mile, nothing practically. That 
shows that the penetration of moisture in that kind of cable does 
not take place rapidly. 

Mr. PATTERSON : Sisal and jute and paper can be classed about 
together in regard to the penetration of water, and it will go much 
more rapidly into cotton. 

Mr. HALL : You made a point about that also, that water 
would not get in any faster than air could get out through any 


ordinary pinhole ; the exchange of air for water would be a very 
slow process. 

Mr. PATTERSON : It depends altogether on where it is. When 
you get a hole on the top of the pipe and the pipe slopes, it will 
penetrate very rapidly, but in the case of a hole on the underside 
of a pipe, it goes very much more slowly. 

Mr. PICKERNELL i The case I had was a case where the cable 
tipped down, going from one side of the manhole to the other. 
It was downhill, and on that account it was only on one side of the 
cable that was injured. The water was very slow in penetrating 
through a pinhole fully as large in diameter as a pin. 

Mr. HALL : Now if we take a vote on the low figure for stating 
capacity, the effect, of course, would be to commit us absolutely to 
the dry cable ; and I would like, as that is an extremely important 
matter, the most radical step which has as yet been taken in cable 
manufacture, to have it considered very carefully. Perhaps you 
have considered it and are ready to commit yourselves. I think 
it would be well to make a motion, and I assume that it is made, 
that we recommend a figure for capacity which shall mean a dry 
core cable ; that without stating now just what the figures should 

Mr. HAYES : I second the motion. 

(Carried unanimously.) 

Mr. HALL : That is the most important step that has been taken 
yet in cable making. 

Mr. HAMILTON : I think our experience so far warrants us in 
taking the step. 

Mr. PATTERSON : There are dry sections which have been in use 
in New York for about four years. 

The conclusion thus reached from knowledge by men who 
recognised their responsibility has been supported by results. Dry 
core cable has been used ever since. It reduced the cost and 
increased the length available for commercial conversation, thus 
rendering the general introduction of underground systems possible. 
The principal factor was the omission of insulating material sug- 
gested by Mr. Patterson, but an important contributory was the 
use of paper due to the experiments of Mr. Sargent and Mr. Barrett. 
Paper provided covering as a separator which was much more 
efficacious than the somewhat solid covering of cotton previously 
used or even the looser braiding subsequently tried. 

The methods of putting the paper covering over the wires varied. 
Cable specifications and cable manufactures were to undergo con- 
siderable changes yet, but these were, in the main, matters of detail 
in manufacture or of increase in the number of conductors in one 
tube, gradually increased from 50 to 300 pairs and larger num- 
bers for smaller conductors, European organisations being perhaps 
in advance in this respect. The conference of 1891, however, 


determined the adoption of the dry core, and in essentials there 
is no change to record in the cable itself. 

The principal changes in the 1891 specifications were : 

Conductors reduced to No. 19 B. and S. gauge ('03589 inch). 

Thickness of insulation modified so that instead of the diameter 
of the individual insulated wire being given, the diameter of a core 
of 52 twisted pairs was specified to be not less than if inch to be 
placed in a pipe not less than if inch internal diameter. Other 
sizes of core and pipe to be in the same proportions. 

Twisted Pairs. Length of twist omitted, and in place thereof 
' The conductors shah 1 be so twisted in pairs that there shall be 
no inductive disturbances between the circuits.' 

Outside Covering. The largest size of pipe (hitherto if inch) 
was increased to 2| inch (5| Ib. per foot), the weights for 2\ inch 
and 2 inch being 4! Ib. and 4^ Ib. per foot respectively. 

Electrostatic Capacity. The Committee contemplated at the 
April meeting an electrostatic capacity of -085 microfarad per 
mile average and -090 maximum but, as the result of practical 
demonstrations available immediately thereafter, the specification 
issued called for -080 average and -085 maximum. In July there 
was a further reduction to -075 average and -080 maximum. To 
make allowance for the filled ends and to permit the manufacture 
of a firmer core the -080 and -085 figures respectively were restored 
at the December meeting. 

Covering for Pipe. The outside covering and asphalt :treat- 
ment of the lead pipe retained only when laid in wooden conduits. 
It was previously considered necessary that this should be done 
generally in order to prevent disintegration of the lead by chemical 
action, but careful inquiry had shown by 1891 that such protection 
was not needed when laid in conduits other than wood. 

An article in the Engineering Magazine of 1843 * describing the 
telegraphs laid on the Blackwall, London, Leeds and Manchester, 
and other railways, states : 

That the minutest changes in the insulation of the wires from 
dampness can be detected by this valuable instrument [the 
' detector '] and corrected by blowing through the pipe a draught 
of dry air from the reservoir. 

In 1858 Captain Drayson of the Royal Artillery and Captain 
Burney of the Royal Engineers obtained a patent (No. 2326) for 
a cable consisting of a conductor varnished and silk covered 

1 Quoted by Mr. A. Watts in a paper on the evolution of underground 
work read before the Manchester Telephone Society, January 15, 1904. 


enclosed in an elastic tube ; and in 1876 a patent (No. 3099) was 
granted to Henry Potts Scott (a communication from Wilson 
Strickler of the United States) for a cable formed of a conductor 
with grooves and furrows and atmospheric insulation, the object 
being to prevent retardation. William E. Prall took out a patent 
in the United States in 1876 (No. 172,495) for a combination of 
one or more telegraph wires or cables with a continuous line of 
airtight enclosing pipes charged with atmospheric air under 

In 1882 Dolbear took out a patent in England (No. 1368) for a 
submarine cable consisting of a conductor surrounded by a spiral 
cord then covered by a waterproofed paper tube enclosed in a gutta- 
percha covering. The object of the construction was to obtain free- 
dom from the retardation due to capacity, and the capacity effect 
was clearly related in the specification as it had been by Wheatstone 
many years previously. 1 An attempt was made to obtain a renewal 
of the English patent in 1896 on the plea that the dry core cable, 
which had by that time come into general use, was Dolbear's 
invention. The petition was heard by the Judicial Committee of 
the Privy Council on March 18, 1896, and failed. 

Amongst others, M. Fortin-Hermann 

sought to combine the advantages of the air line with those of the 
underground cable. . . . The conductor is inserted in small wood 
cylinders touching one another by the ends, and thus forming a 
chaplet or chain, covered by sheet lead or inserted in a lead pipe. 2 

Glass beads were also tried in place of the wood cylinders, but 
neither form was sufficiently practicable to survive. The Fortin- 
Hermann cable, however, was an early example of the effort to 
attain low capacity and was used in the underground portion of 
the Paris-Brussels line (p. 430). 

Individual invention may claim some credit on minor points 
of detail, but the dry core system as a whole is not the subject of 
patent. Its utility was recognised and its adoption was recom- 
mended by a committee of experts after experience had demon- 
strated that confidence could be placed in the mechanical integrity 
of a lead tube. 

1 Chapter xx. p. 261. 2 Electrician, February 21, 1885, p. 303. 



DURING the first ten years of telephone service numerous systems 
were in practical operation. Some managers were responsible for 
the design of a complete ' system/ others were content to invent 
a switchboard, but most had some detail which in their judgment 
made their own exchange a model suitable to be copied by every- 
body else. Space does not permit giving particulars of all the 
various modifications which were put in practice, but reference will 
be made to the principal ones. 

From the construction standpoint there were two main systems 
' one wire ' and ' two wire.' These terms might be understood as 
earth circuit and metallic return respectively, so that it is advisable 
to add that they are not here so used. The ' one wire ' was a 
direct line to the exchange, through which line was conveyed instruc- 
tions to the exchange operator as well as conversation between 
subscribers. ' Two wire ' meant two lines to the exchange, one 
direct for conversation with other subscribers, the second (usually 
common to several stations) for conveying instructions to the 
exchange operator. 

From the standpoint of the calling subscriber the choice of 
systems may also be taken as two : 

(a) To call the exchange by the mechanical operation of a 
magneto generator or by pressing the button of a battery circuit. 

(b) To communicate directly with the operator through the 

A combination of these two, whereby the indicator operated by 
magneto or battery connected the line by means of a local circuit 
to the listening operator's telephone, may be mentioned, but its use 
was not sufficiently extensive to be included in the above. The 
earliest device of this kind is to be found in the United States patent 
of John A. McCoy, No. 278,351, where the shutter of an indicator 
is part of the line circuit and on being operated breaks an earth 
contact and makes contact with the operator's telephone circuit. 



In other words [as the patentee says] my invention consists in pro- 
viding an automatic switch to connect the central office telephone 
with that of the person signalling, the said switch to be operated 
at the same time that the central office is signalled. 1 

A similar purpose is served in the system described in Scribner's 
United States patent No. 278,367? 

A further classification is needed for indicating the conclusion of 
the conversation or the desire for the disconnection of the lines : 

(c) Operating magneto or battery bell in the same manner as 
for calling. 

(d) Operating magneto or battery so as to send a different 
current from that used to call. 

(e) By communicating directly with the operator through the 

Taking the service as a whole and combining the calling and 
clearing, (a) and (c) represent the more general magneto (or battery) 

(a) and (d) represent the magneto (or battery) ' Ring through ' 
system, and (b) and (e) the Law (or call wire) system. 


In the magneto system there is a direct line from subscriber 
to exchange. At the subscriber's office a magneto generator with 
bell to send or receive signals, and talking instruments ; at the 
exchange an indicator and spring jack. 

The diagram fig. 107 represents the lines of two subscribers to an 
exchange : A is the magneto at the office of subscriber No. i, B is 
the spring jack and C the indicator of No. I subscriber at 
the exchange, D is the magneto at the office of subscriber No. 2, 
E is the spring jack and F the indicator of subscriber No. 2 at the 

The handle of the magneto generator being turned, a current 
is generated which drops the indicator shutter, and the operator 
is thereby apprised that attention is required. The operator, 
having ascertained the number required, completes the connection. 
The lines when connected may be represented by the diagram 
fig. 108 : where G and H are plugs and I a clearing-out indicator 
operated by the same current as the calling indicators C and F. 
A having called and desired connection with D, the operator calls 
D by means of devices not illustrated on the diagram and then 

1 U.S. specification, No. 278,351. Dated May 29, 1883 (application filed 
April 5, 1880). 

* Dated May 29, 1883 (application filed November 6, 1880). 


connects them for conversation as shown. The request to dis- 
connect is communicated to the operator by either subscriber 
turning the handle of his magneto exactly as in the original call. 
The clearing-out drop I falls and the operator withdraws the plugs 
G and H, when the lines assume the condition shown in the dia- 
gram fig. 107. 

To carry out this method satisfactorily, it is required that a 
subscriber should only send a current over the line for the purpose 
of calling the exchange or for disconnecting. It was with difficulty 
that subscribers were brought to understand that attention to such 
a limitation was essential. It was not unusual for two subscribers 

FIG. 107. 

FIG. 108. 

(A and B) to hold a prolonged conversation, and for A perhaps to 
leave the instrument to acquire information for his correspondent 
B. After a time B, becoming impatient, would turn his magneto 
handle under the impression that he would thereby bring A again 
to the instrument. The result, however, was of course to throw 
the clearing-out indicator, which was the signal for disconnection. 
The uncertainty was considerable, and many authorities had 
not acquired sufficient information to realise the importance of 
insisting upon the accurate carrying out of instructions or placing 
upon the subscriber the penalty of any infraction. To quote one 
such set of instructions on this point (Melbourne, Australia, 1883) : 
' When connected do not again use your bell except to ring off. 
Should you do so you will be immediately disconnected.' In 
many exchanges there was not the courage to instruct the subscriber 
by carrying out this salutary rule. In consequence there was 


uncertainty, and operators were sometimes accustomed to ' listen 
in ' and make sure before disconnecting. 

The ring-through system avoided this uncertainty and in 
addition provided what was regarded by some as a valuable facility 
in permitting two subscribers to be connected for an indefinite 
period. The advantages of the ' ring-through ' were clearly limited 
to a single switchboard. If more than one switchboard were con- 
cerned in the call, the first operator having connected the calling 
subscriber would have connected him, not to the correspondent he 
required, but to another operator. The second operator would 
connect the line to the called subscriber, and the subscriber origi- 
nating the call would ring. 

The encouragement to two subscribers to leave their lines con- 
nected for an unnecessary period was a mistaken policy in a single 
exchange. Other subscribers might call for one or both, and their 
continued engagement would be a disadvantage. But, in the case 
of two exchanges, it would be the more serious because trunk (or 
junction) wires would be employed. The number of junction 
wires between exchanges is in proportion to the traffic, and the 
number must be increased if the durations of the connections are 

One other respect in which the ring-through system commended 
itself to some telephone engineers was that it placed the 
responsibility for delay in answering calls upon the subscriber. A 
calling subscriber awaiting the answer to his call attributed to the 
exchange operator the responsibility for any delay. So much of 
it as was contributed by the subscriber was in no sense the operator's 
fault, but in the absence of any information to the contrary the 
subscriber placed the responsibility for all delay on the exchange 
service. To get rid of some of this responsibility was attractive, 
and the ring-through system consequently had its adherents. 

An example of the ring-through system with batteries was 
that at Manchester described by the designer, Mr. Poole, in the 
first edition of his ' Telephone Handbook,' l also in the ' Manual of 
Telephony ' by Preece and Maier. Batteries rendered the system 
simple, since it was only necessary to use one pole for calling and 
the other for clearing. 

An example of the ring-through system with magneto was 
that in operation at Omaha, Nebraska. The magneto was corn- 
mutated and an alternating or direct current could be sent at will. 
The calling indicator at the exchange was of the ordinary type 
operated by alternating current. The clearing-out indicator was 
specially designed so as to operate only by a direct current. It was 

1 Practical Telephone Handbook, Poole, 1891 edition, p. 168. 


called a ' galvanometer drop/ a name which is fairly descriptive 
of its design. But a still earlier plan is that of C. A. Hussey, 1 in 
which also a commutated magneto was employed. 


The Law system derived its name from the fact that it was the 
system used by the Law Telegraph Company of New York in the 
operation of its dial telegraph service prior to the introduction of 
the telephone, and was retained by that company when the dial 
instruments were superseded by telephones. 2 It was one of the 
earliest exchange methods in use. A circuit calling wire was used 
in Chicago as described in detail in Chapter IX, and the multiple 
board of Firman 3 had no indicators. For calling and disconnecting 
the ' district ' circuit wire was used. Though there is evidence of 
use by others and the suggestion by the district system is obvious, 
priority of publication must be given to the Law Company for the 
application of the idea to telephone exchange service by reason of 
the grant of a patent 4 to Frank Shaw of New York, Assignor to 
Law Telegraph Company of same place. Priority of invention was 
claimed by Firman, who was granted a patent (No. 328,305) on 
October 13, 1885 (application filed January 16, 1880), antedating 
that of Shaw. Shaw's specification, following the usual custom, 
sets forth the drawbacks of prior systems for the purpose of 
indicating the superiority of the method which he claimed to have 

The object of the invention is 

to enable any one of a large number of subscribers to what is com- 
monly known as a ' telephone intercommunication exchange, district 
or central office system ' to be quickly, easily, and without confusion 
put in private and direct communication at any time with any other 
subscriber. 4 

He describes as the usual method the employment of a line 
from the central office to the office of the subscriber, over which 
direct line are conveyed the signals or instructions to the operator 
as well as the conversations of the subscribers, and proceeds : 

This system is open to several serious objections as follows : 
First, it frequently leads to confusion, because many subscribers 
sometimes do, and all may, at one time call or signal to the central 
office for connections, which cannot be properly received and executed 

1 U.S. specification, No. 247,359. Dated September 20, 1881 (application 
filed April 9, 1881). z Chapter ix. p. 86. 3 Chapter xix. p. 213. 

4 U.S. specification. No. 220,874, October 21, 1879. 


simultaneously ; second, it requires a large number of operators 
for such emergencies, who at other times stand about idle and 
unemployed ; third, it consequently leads to frequent mistakes 
and delays ; fourth, it requires considerable space and a large 
amount of apparatus. 

Therefore, while this system seems simple, owing to the use of 
a single wire for calling the central office and communicating with 
a subscriber, and the same wires for communications between 
subscribers, it is in fact complicated in operation. 

I overcome these objections by employing two wires and circuits 
one for connection or disconnection calls or signals, and the other 
for private communication between subscribers constructed and 
operated in the manner hereinafter set forth. 1 

The next paragraph in the specification has nothing to do with 
the system and is not referred to in the claims, but it is of interest 
as a contemporary record of the inconvenience resulting from the 
use of subscribers' names instead of numbers : 

It has been found that where there is a considerable number of 
subscribers, where firm-names are composed of several individual 
names, and especially where more than one subscriber has the same 
name, confusion and mistakes are likely to occur both at the central 
office and at subscribers' stations. I avoid these difficulties by 
numbering each station and printing a schedule or list of sub- 
scribers, with the name or title of the subscriber opposite the 
number of his station, which list or schedule is posted up near the 
instruments at each station and office. The calls and notices for 
connection are then all conducted without error by the use of 
numbers to designate the subscriber, which fixes absolutely the 
actual parties desiring to communicate or communicating. 1 

In describing his invention he says : 

From each subscriber's station I run two wires to a common 
central office, one of which wires I preferably run to no other station 
or stations, although several stations may be located thereon ; and 
the other wire I run to as many other subscribers' stations as the 
amount of business done by them warrants. The first is the private, 
and the last is the call or signal wire. These wires are provided 
with a battery or other source of electricity and with a suitable 
switchboard and instruments at the common central office, and 
suitable instruments and batteries or other source of electricity at 
all the stations. Then any one subscriber can at any time obtain 
private and direct telephonic communication with any other 
subscriber by requesting the common central office, by means of the 
last-mentioned call or signal wire, to connect together the private 
wires of the two subscriber-stations, and afterward, having finished 

* U.S. specification, No. 220,874, October 21, 1879. 


their communications, can signal in like manner the common central 
station to disconnect said private lines, so that either may be in 
readiness to be connected at any moment with the private line of any 
other station. 1 

The advantages of the circuit wire system are thus set 
forth : 

The call or signal wire being common to a considerable number 
of subscribers' stations and the instruments there, the result is that 
a subscriber going to his instruments is able to learn whether or 
not any other subscriber is signalling, and to await his turn, or until 
the call or signal wire is not in use before attempting to signal. The 
result is that the subscribers, as it were, form themselves in cue, 
each taking his place in succession. 1 

The method of operation in the exchanges, as described in the 
specification, contemplates the employment of two operators, one 
to listen, the other to do the switching. Such a method was not 
limited to the Law system and was not generally employed in Law 
exchanges, switchboards for which were simple. Though the work 
was divided in the first instance, it became general for the operator 
on the circuit wire to carry through the actual connections as well as 
receive the instructions. 

At the office end of the signal wire I place a receiving operator, 
who sits with the receiving instrument or telephone constantly at 
or near his ear, ready to receive the name or number of any sub- 
scriber desiring to communicate and of the correspondent with whom 
he desires to be placed in communication. The names or numbers 
so received he repeats so as to be heard by the subscriber calling 
through the transmitting-instrument in front of him, and at the 
same time by another operator at the switchboard in the central 
office, who instantly connects the two private wires and subscribers 
so indicated. As he does so, the operator at the switchboard notifies 
both subscribers by means of a bell in circuit on the wire. The same 
course is pursued when the conversation is concluded and discon- 
nection is desired, except as to the ringing of the bell. 1 

The descriptions are as follows : 

Fig. [109] is a plan of my system for connecting a central office 
and its subscribers. Fig. [no] is a plan of my system for connect- 
ing the subscribers who belong to one central office with those who 
belong to another central office. Fig. [in] is a plan showing two 
subscribers' stations and central office with instruments and receiving 
operator in position. 

In fig. [109] X represents a central office or station, and a, b, c, 
represent a given number of subscribers' stations, each of which 

1 U.S. specification, No. 220,874, October 21, 1879. 



is connected with the common central office or station by two wires, 
the first of which is marked W, and the second a 2 b 2 c z , respectively 
and also W 1 in common. 1 

The method of calling in a single exchange is described but 
need not be quoted. There 
is some interest, however, 
in the reference to numer- 
ous exchanges : 

In cases where dis- 
tances are great I establish 
two or more common cen- 
tral stations, converging at 
each the wires from all 
stations nearest to it and 
connecting all the common 
central stations together 
by as many wires as may 
be required. 

Now, should a wish to 

communicate with the subscriber of any other central office, as d, 
(fig. [no]-), he makes known the fact to the operator at X by means 
of the wire W, who immediately connects the wire a 2 with one of 
the wires A 1 , A 2 , or A 3 , and instructs the operator at the station 
Z by means of another of the wires A 1 , A 2 , or A 3 , used as a call or 
signal wire between central stations (two or more) to connect the 
wire d z with the wire A 1 , A 2 , or A 3 , first mentioned. 

FIG. 109. Law System Single Central 
Office (Shaw's patent). 

FIG. no. Law System Two Central Offices (Shaw's patent). 

The disconnection is accomplished either by the subscriber at 
station a notifying the operator at the central station X, and he 
notifying the operator at the other central station Z by the same 
wires as before, or by the subscriber at the station d notifying the 
operator at Z, and he notifying the operator at X. 1 

The Law system was sometimes called the ' Call Wire System,' 
the ' Two Wire System,' and in Great Britain the ' Mann System.' 

1 U.S. specification, No. 220,874, October 21, 1879. 


The latter name was given to it under the impression that Mr. Mann 
was the first inventor of an improvement in the running of the 

circuit wire by taking spurs into the subscribers' offices instead of 
looping in and out again. The connection by derivation reduced 
the inconvenience from the failure of the circuit wire, but, as was 
shown in the course of a controversy on the subject in the London 


Electrical Review of 1891 (vol. 29), the derivation method was 
common in the Law exchanges of the United States. 

In 1887 the exchange service had been in operation ten years. 
The problems which those ten years had presented were of the most 
varied character. In the conduct of this service, which gave to 
the public an entirely new means of communication, the providers 
had to contend with conditions in which electrical and mechanical 
problems were intermingled. Difficult as some of these problems 
were, they were at least capable of settlement in the light of scien- 
tific knowledge. But the human element required more study, and 
the human element was at the beginning, at the end, and at the 
middle of the line. The last mentioned was the operator at the 
exchange, a servant of the company available for instruction and 
subservient to discipline. But the subscriber who initiated the 
call, and the other subscriber who received it, were very important 
contributors to the satisfactory completion of a connection. Upon 
these contributors the company relied for its revenue. Whilst 
discipline was desirable it could not be enforced. The subscriber 
had to be studied and, so far as possible, to be educated. 

The systems in operation afforded material for comparison, 
but the absence of uniformity was a drawback. The time had 
come to weigh experience and to select the best under all the 



TELEPHONE exchanges had grown and flourished, using different 
methods and materials whilst aiming at the same result. In Europe 
the companies or administrations carrying on the service had no 
relationship to one another, and there was little disposition to 
exchange information or suggestions. In the United States there 
were many exchange companies, but all had a common interest 
in their relationship to the parent company under whose licenses 
they worked. In the United States people with a common interest 
are accustomed to indulge in periodical conventions. The annual 
meetings of the National' Telephone Exchange Association were 
called ' conventions,' and from 1880 onwards these meetings had 
afforded valuable opportunities for comparing experiences and 
obtaining information. These opportunities were taken advantage 
of by the progressive managers of telephone companies, and much 
information was freely exchanged. In some respects results were so 
clear as to determine the line of further work. But there was no 
provision for reaching definite conclusions by evidence arguments 
and votes, and no effective authority for enforcing any such general 
conclusions as might be reached. 

Variety in methods and material was a natural consequence of 
different minds being responsible for the inauguration and carrying 
on of the exchanges. Whilst the parent company in the United 
States, for instance, was ready to advise or to help, it gave no definite 
instructions or even general recommendations to its licensees as to 
the systems which they should adopt. There is no evidence that 
this attitude was the result of any preconceived plan, but there 
can be no doubt of its advantage. The exercise of any dogmatic 
authority in a new enterprise is a mistake. There was no experience 
of telephone exchange service upon which reasoned decisions could 
be based. The wisdom of leaving a free choice in methods has been 
demonstrated in one instance by the fact that the multiple principle, 



a feature in switching mechanisms destined to survive, was the 
product of a manufacturing company not connected with the Bell 
Company. It was the obvious advantages of the multiple switch- 
board (though not at that time fully developed) and the clear evidence 
of the mastery which its designers had obtained over exchange 
problems, as well as the patents they held, that induced the American 
Bell Company to acquire the Western Union Company's interest 
in the Western Electric Manufacturing Company in 1881. 

As the business progressed it became evident that there was 
a limit to the advantages which would result from having each 
exchange run on the plan which might appeal to its own manager. 
It seems to have been recognised that excellence could not be deter- 
mined merely by individual opinion ; that there must be some one 
system amongst the many in operation which was better than the 
others ; or that some combination might be obtained which would 
be superior to all. Opportunity had been given for obtaining 
information regarding the effectiveness, the cost, the durability, 
and the popularity of different methods and material. The time 
had been sufficient to demonstrate also that experiments on the 
large scale in the live exchange were very costly in cash outlay and 
very disturbing to the service. Moreover, the advantages of uni- 
formity in operating methods might be expected to appeal to the 
American Bell Telephone Company, for from the first the controlling 
spirits of that company, had contemplated a national system, and in 
pursuance of that object had retained an interest in all trunk lines 
as well as a substantial interest in local exchanges. By these means 
they were able to exert influence upon the engineering and operating 
methods of the various exchanges which enabled them to advise 
with some authority what systems should be used. 

It was with the object of obtaining accurate information and 
determining the most suitable policy for the immediate future 
that it was decided to hold a conference of representative men to 
discuss central office methods. The credit for calling the conference 
was ascribed by Mr. W. D. Sargent during its sittings to Mr. Theo. N. 
Vail and Mr. E. J. Hall, but the latter stated that this particular 
conference was held at the earnest request of Mr. Hudson, then the 
Vice-President and General Manager of the American Bell Telephone 
Company, who was anxious that everything should be done to bring 
out all the information available for the development of the business. 
The proceedings of the conference are contained in a printed volume 
of 255 pages which has never been published. These proceedings, in 
combination with those of the cable conference held in the same year, 1 
may be considered as placing the arts of telephone exchange 
engineering and management arts, perhaps, of greater complexity 

1 Chapter xxii. 

x 2 


and requiring more careful thought and more freedom from prejudice 
than any other practical applications of science so far attained- 
for the first time on a scientific basis. 

The conference was an early example of co-operative effort 
or ' team work ' such as has become by this time a commonplace 
of telephone practice. Though representative men, not all the 
members grasped the principles underlying the details of apparatus 
or service ; but the members exchanged views without restraint, 
and the imperfect knowledge of some only served to afford oppor- 
tunities for definite demonstration by others, so that all points were 
brought out with clearness and precision. 

Individual opinion was still of value, but henceforth it must be 
based on demonstrable experience or sound reasoning which should 
appeal to other minds initiated in the mysteries but open to 
conviction. Thus, in the United States at any rate, telephone 
engineering was emancipated from the rule of thumb or the personal 
predilections of individuals and placed upon a more scientific basis 
so far as the experience at that date permitted. 

A synopsis of eight pages, prepared by Mr. Lockwood, summarises 
the conclusions of the conference. The reasons for the summary 
are thus stated : 

The full report of the proceedings, while of great value, 
is so long, that much time, careful research, and close attention 
would often be necessary to ascertain the opinion of the conference 
expressed or implied on any question. It has, therefore, been 
thought well to prepare and append a synopsis formulating the 
absolute conclusions reached, and also those points of agreement 
which the drift of the discussions indicated as representing sub- 
stantially the sense of the conference. 

' The art of blotting ' is said by Goldsmith, on the authority 
of ' an eminent critic,' to be ' the most difficult of all arts.' He 
was himself engaged upon abridgments, which he said ' are generally 
more tedious than the works from which they pretend to relieve 
us, and they have effectually embarrassed that road which they 
laboured to shorten.' 

This was, however, only because the art of blotting had been 
' usually practised by those who found themselves unable to write.' 
The synopsis above referred to is an exception to Goldsmith's 
generalisation, having been compiled by one not only ' able to write ' 
but also familiar with every feature of the work covered. It was 
probably prepared in order to ' shorten the labours ' of Mr. Hudson 
and the executive officers of the company in acquiring a knowledge 
of the practical conclusions of the conference, and to such practical 
conclusions it is severely limited. 


Since space prevents the publication of the full report, and 
abridgment becomes essential, I must acknowledge a temptation 
to incorporate the synopsis. But ' to attain the greatest number 
of advantages with the fewest inconveniences/ which, to quote 
Goldsmith again, ' is all that can be attained in an abridgment 
the very name of which implies imperfection,' I have concluded 
that it will be better to draw upon both the full report and the 
synopsis, making such extracts therefrom as will serve to record the 
conclusions reached and the reasons for them, as well as to give 
some insight into the conditions of the time. 

The peculiarity of the Telephone Exchange Service, compared 
with other public services or methods of communication, lies in 
the combination of individuals and apparatus. In transport and 
in prior methods of transferring intelligence, the public perform 
a merely passive part. The telephone required of the public an 
active participation. 

The individuals engaged are of two kinds the subscribers and 
the operators. The operators, and the apparatus which they 
manipulate, are for the service of the subscribers, and, therefore, 
in abridging the proceedings of the conference, it will be convenient 
to select such points as deal with them in the following order : 

1. General. 

2. Subscribers. 

3. Operators and other servants of the exchange authority. 

4. Apparatus. 

i. General. The conference 1 was held in New York on 
December 19, 20, and 21, 1887. 

In the course of his introductory address Mr. Hall said : 

For the past ten years the telephone service has presented an 
unending series of problems, and to-day the complicated condi- 
tions which surround our large exchanges present difficulties still 
to be surmounted which will tax to the utmost our united energies. 
The very satisfactory results of our recent conference on telephone 
cables lead us to hope that this meeting may accomplish something 
of equal importance for the switchboard. It is only by attacking 
these difficulties in detail that we can cope with them successfully ; 
and, while there are many other subjects of equal interest, let us 
confine the work of this meeting as closely as possible to the central 
office and its apparatus. 

From the standpoint of a general manager the first problem 
connected with central office apparatus is to find a place to put it. 
The stuffy attics of early days, with a little cupola reached by a 

1 The members were : E. J. Hall (in the chair), E. M. Barton, T. D. 
Lockwood, A. S. Hibbard, W. D. Sargent, J. A. Seely, C. E. Scribner, 
Henry Metzger, F. G. Beach, B. E. Sunny, C. H. Wilson, Wiley W. Smith, 
J. P. Davis, W. J. Denver, and I. H. Farnham. 


ladder through a tangled mass of wires, can no longer be the lodging- 
place of the modern telephone exchange. At best, it was but a 
boarding house, with all its discomforts and uncertainties. Now 
we must have homes which we can build or arrange to meet our 
varied and complicated needs. 

Every large exchange should be located in a building owned by 
the company, and that building should be located and arranged 
after a careful study of the territory to be served and a most 
liberal estimate of the future needs of the business for which we are 

Ample space, light and air must be provided for the operat- 
ing department, and convenient toilet and lunch rooms for the 

Overhead and underground wires must be' brought in in such 
a manner as to be fully accessible ; suitable testing apparatus must 
be provided, and competent electricians and office managers em- 
ployed to supervise every detail and keep the machinery up to its 
highest efficiency. 

As to the desirability of these things there can be no question, but 
their practical necessity should be impressed upon the directors of 
our companies so that they may not make the costly mistake of 
equipping any large exchange office in other than permanent fire- 
proof quarters. 

In these quarters we can put no flimsy makeshifts in the way 
of apparatus. We must require of the manufacturers strong and 
durable materials, put together substantially and accurately. 
What those materials shall be in what way they shall be combined 
and how the various parts shall be arranged to equip the ideal central 
office are questions for us now to consider and decide. 

As a basis for discussion, Mr. Lockwood presented a memorandum 
summarising previous experience and formulating certain questions 
for decision. The thoroughness with which the problem was 
attacked may be gathered from the introductory definition of the 
purpose for which the switchboard was required. 

Switchboards. By this term, for the purposes of this conference, 
we are to understand the entire apparatus of the central station, 
whereby any two subscribers are enabled to converse with one 

To bring the subject intelligently before us, the first thing to 
do is to consider and formulate the purpose for which the central 
station switchboard is required. 

A hasty reply to the question, ' What is the function of the 
switchboard ? ' would be, and often has been : ' To connect the 
ends of any two subscribers' lines together so that they can com- 
municate directly with one another just as if they were united by 
the same wire.' 

This looks like a simple operation, but we know that, while 


this is a true statement of the final result, it is not the whole truth. 
There are a variety of sub-operations, all tending to this end. The 
several steps, as I understand them, are as follows : 

To receive the call signal from the subscriber. 

To respond by answering, and then to receive the order. 

To ascertain whether the subscriber's line wanted is in use. 

If so, to notify the calling subscriber of the fact, and so for the 
present to end the operation. 

If not, to send the call signal over the line of the subscriber 

To ascertain if he answers his bell. 

If not, to notify calling subscriber that No. 2 does not answer, 
and thus conclude the operation for the present. 

If so, to connect the two lines, removing the ground terminal 
of both, and bring subscribers together, if necessary. 

To receive disconnecting signal. 

To disconnect. 
To perform all these functions we want : 

Call-receiving appliances. 

Call-sending appliances. 

Telephones to receive and forward orders. 

A switchboard proper to connect wires together. 

An appliance to connect and disconnect the telephones with 
and from any desired line. 

Some appliance, mechanical or otherwise, to find out whether 
lines are in use. 

Appliances to receive disconnecting signals. 

These operations and appliances to perform them are not 
common to all central offices, but the five operations of call receiving, 
call sending, finding whether line wanted be in use, and connecting 
and disconnecting in some form, are inseparable from the work of 
the central office. 

We have now, I think, to consider the principal switchboards 
which have been in use, especially the multiple switchboard in 
which all others seem to have culminated. 

First. With respect to efficiency. 

Second. With respect to economy. 

Third. With respect to permanency. 

On the lines of this introductory definition Mr. Lockwood formu- 
lated twenty-two questions for the consideration of the conference. 
Each question had a definite and practical bearing and was considered 
by the members in detail, it being recognised that 

The problem of the switchboard and of central office work is 
not one which can be worked out by electricians alone, however 
capable. It is more mechanical than electrical, and it is one in 
which the experimental knowledge of men who have practically 
operated large exchanges also will count materially. Electrical 

and mechanical skill and experience must therefore together consider 
the subject. 

Attention was given not only to definite practical questions, 
but also to the avoidance of possible dangers of a more general 
character. One of the questions, for example, related to the 
practicability of 

setting aside all apparatus which as a whole is now used, and by 
availing ourselves of the universal stock of knowledge to devise and 
construct something which shall be more efficient, more economical, 
or more endurable, and whether in such an enterprise it would not 
be judicious to associate with known switchboard experts one or 
more persons who, while electricians or electro mechanics of acknow- 
ledged skill, have had no special views upon switchboards or switch- 
ing organisations, for the purpose of obtaining absolutely new ideas 
unfettered by previous practice or conditions. 

In introducing this question, Mr. Lockwood said that it was 
based upon views which Mr. Theodore N. Vail had frequently 
expressed to him, that nearly all switchboard experts and telephone 
men of that era were working in ruts, 

that we have got accustomed to certain things, and we cannot 
see our way or feel our way or see any other possible way out of it 
than to go to a certain extent as our predecessors have gone, and 
he (Mr. Vail) thought it would be a good thing if we could find 
somebody somewhere who was a first-class man in every other 
respect but who had not devoted his attention to switchboards. 

A sub-committee had been formed, consisting of Mr. Scribner, 
Mr. Hibbard, and Mr. Wilson, to investigate a specific question ; but, 
as the sub-committee was to report on the following morning, it 
was not practicable to add the suggested new member, but the 
chairman thought it was well to have Mr. Lockwood's question 
on record to consider. This conference and its successors, com- 
posed as they were of practical men in various branches of the 
service, seem to have been so constituted as to avoid the danger 
of the ' one idea ' or the getting into a groove. 

2. Subscribers. --The education of the subscriber in the use of 
the telephone was agreed upon as a very desirable measure. Mr. 
Sargent estimated that fully two-thirds of the troubles experienced 
were directly traceable to the users. Mr.' Hall explained that in 
Buffalo the ' chief canvasser or contract agent ' undertook the 
work of educating the users. 

When he goes to people on this business he does not talk any- 
thing except the good of the subscriber, the defects of the service. 


and expressing the desire of the company to be sure that its sub- 
scribers are receiving the service for which they are paying. . . . 
Of course, to send around a man who did not possess a good deal of 
tact and judgment would simply be to cause trouble rather than 
to remove it. 

We have found another incidental advantage in Buffalo in that 
missionary work, that a great many complaints, more or less well 
grounded, have been brought to the attention of the company that 
would not have been received in any other way, and a great many 
mistaken ideas of subscribers have been corrected ; and the whole 
effect of the thing has been, I think, to impress them with the belief 
that the company are sincerely interested in giving them good 
service and took pains to ascertain whether the service was satis- 
factory to them, and if not, what they would suggest in the way 
of remedying it ; and then, hearing their suggestions, the company 
has an opportunity to point out to them where the fault lies with 
them, or if the suggestions are good ones, and some of them have 
been, to follow them out and correct abuses in the service. 

One of the objections to the Law system was the difficulty 
of educating the subscriber in its use ; and another respect in which 
systems and subscribers were considered, was to put the proposition 
as to whether it would be easier to change from the Law to the more 
general magneto system or 'vice versa. It was recognised that 
subscribers objected to any change. This was alluded to by Mr. 
Smith, who illustrated his views with a story then current : 

Suppose I should find the Law system most desirable. As 
Mr. Denver intimated a moment ago, it would create a great 
disturbance in Boston, I know it would in my city, to attempt to 
change. No matter how desirable I might think it to be, or my 
directors might think it to be, I should hesitate to attempt to 
change ; I should expect to find myself in the position of the boy 
that I read about not long ago, who was sent upstairs to bed, during 
the prevalence of a thunderstorm, against his vigorous protests. 
He was finally persuaded to go by telling him that God would be 
there with him, and it would be all right. He did go ; his mother 
kissed him good-night, and he went upstairs. Then came an 
exceedingly loud clap of thunder, and the next thing there was a 
little boy heard at the top of the stairs, saying, ' Mamma, mamma, 
you come upstairs and stay with God awhile, I am coming down/ 
I think that is about the situation. 

Systematic education of the subscriber had been advocated 
before the 1887 conference. The remarks of Mr. Fay at the 1884 
meeting of the National Telephone Exchange Association give 
one instance. He was referring to the records of incompleted 
connections : 


I think that probably ninety out of every hundred of uncom- 
pleted connections result from busy wires alone, and that the 
uncompleted connections are particularly owing to the fact that some 
wires are very busy. For instance the heavy railroad wires can 
generally be charged with half of the uncompleted connections. 
In other words, they are busy all the time, and it is impossible for a 
man to call for a wire and get it, except by taking his turn. The 
drag of the subscriber on the exchange, however, is because he is 
slow in answering, and careless about doing his part of the work, 
ignorant of the way in which the machine works, hasty in making 
his calls, indistinct in his speech, or for some other similar reason, 
and he becomes an unmitigated nuisance in every way except as a 
payer of rentals. We have lately undertaken the missionary work 
of converting him and of making him more amenable to reason, 
of making him feel that he ought to know something about his 
machine, how to work it and how to get the best results out of it. 
We have found that the heaviest users we have make the smallest 
number of complaints in reference to the service of the telephone. 
They know how to use the telephone, and if they get into an excep- 
tional condition, where the telephone will network satisfactorily, 
they know how to get out of it. As I say, we have gone into the 
missionary service in that respect, and for that purpose we have 
four men going around the city of Chicago all the time teaching the 
subscribers how to operate the telephone. That has certainly been 
of great assistance to us. 1 

A question of great importance at the 1887 conference was : 
How much should the subscribers be called upon to do ? Mr. 
Scribner pointed out that this determined the number of connections 
that should be required of an operator in a given time and the cost 
of the apparatus by which the work was done. 

The least that an operator can be required to do is to get the 
number of the line wanted and to connect the two lines together. 
The operator may do more than this. She may call the wanted 
subscriber with a ring of the bell. To ascertain the number, to 
connect the lines together, to ring the wanted bell, and to wait 
in the line until the bell is answered and until the two subscribers 
get to talking, is the extreme of the ordinary requirements of an 
operator. But sometimes the line wanted will be found busy. In 
this case the calling subscriber must be informed of the fact. But 
in that case it may be left for the calling subscriber to call again at 
his convenience, or the operator may be required to remember the 
number of the subscriber calling, and to watch for the busy line, 
and when it is disengaged to get the two subscribers to talking. 
This represents about the extreme of requirements for connections 
other than toll-line connections. 

Mr. Hall expressed as his judgment ' that the function of the 
1 National Telephone Exchange Association Report, 1884, p. 66. 


telephone exchange is to receive calls from its customers, call up 
whoever is wanted, inform them who wants them, see that the lines 
are connected together and the parties properly talking before they 
leave it. That is the extreme case that you have stated, and I 
believe that anything short of that is falling that much short of the 
full duty of the exchange.' 

The first ' conclusion indicated by drift of discussion ' recorded 
in the synopsis is that 

It is the duty and obligation of a central station when a call is 
received, to follow the call through and to bring the subscribers 
actually into communication with one another, not merely to receive 
the call and ring the subscriber desired. 

It is, however, recognised that there are many diverse conditions 
in the several exchanges, and that methods must differ to suit 
the varying conditions which should be studied. While in one 
exchange the proper procedure, where a number of calls closely 
follow one another, is to answer each, ring the subscriber desired, 
and connect the lines, returning to each a moment later to make 
sure that the two subscribers are in communication ; it is also 
desirable, where time will permit, that a connection shall be fol- 
lowed closely by the operators until the subscribers have received 
the service. 

It will be seen that systems which threw the work in part upon 
the subscribers, such as the Law system and the ring-through system, 
were ruled out not only on account of such particular defects as 
were observable in practice, 1 but also on the general ground that 
the service which they afforded did not attain the standard con- 
sidered to be necessary. 

Some economy might be obtainable by dividing the work between 
subscriber and operator, but that economy would be dearly purchased 
at the price of an imperfect or unsatisfactory service. The 
conference placed the standard of service high, and indicated that 
it was the duty of the exchange to undertake any work which 
might be required to attain that standard. It was an important 
pronouncement, for good service dominates all other considerations 

The relations of the exchange authorities to the subscribers 
and the manner of communicating with them were carefully con- 
sidered, even to such points as the difference between ' won't 
answer ' and ' don't answer.' Mr. Smith expressed the objection 
to the former. The fact was that the subscriber did not answer. 
' I have spent a great deal of time to get my, operators to use the 
term " He does not answer." That conveys an entirely different 

1 Chapter xxiii. 


3. Operators and other servants of the company. The need of 
strict discipline in order to get an efficient service was insisted upon. 
Mr. Smith said : 

I have always instructed my operators as to the exact manner 
in which they should convey the desired information to subscribers. 
There are but few things they have to thus convey to them : first, 
whether the line is, or is not in use ; whether the line is out of order, 
and one or two other things ; and these are given in the same 
language by all of our operators, and we hold them to it strictly. 
We have two assistant chief operators in our Kansas City exchange, 
while We have from thirty-five to forty operators on duty all the 
while. Those two assistants are really floor walkers, who are up 
and down the room all the while, watching operators in their conduct 
of the service. As to the matter of ringing also, they are instructed 
to give only short rings. We believe that a short ring excites the 
curiosity of the subscriber, and that he will answer it quicker than 
he would a long one, which irritates him ; observation has proven 
that to be the result. 

We answer by asking : ' What number ? ' The number being 
given, we repeat the number so that the subscriber may know that 
his order has been received and correct it if an error has been made 
in understanding it. If the line is in use, we say, ' They are talking.' 
If the line is crossed or otherwise not available by reason of trouble, 
we say, ' The line is out of order.' If the line is ready for use, she 
simply calls and passes to the next drop that is down, and so on, 
until she has answered all the calls before her, and then comes 
back over them to see if they are talking. If any of them have 
been connected long enough, as she may think, to be through, she 
tests by asking, ' Did you get them ? ' provided she hears no talking. 

Mr. LOCKWOOD : She listens first ? 

Mr. SMITH : She listens first, and, hearing no conversation, 
she asks, ' Did you get them ? ' We used to ask ' Are you through ? ' 
following the example given us, but we found that irritated the 
subscriber, as the operator would frequently meet with apart}' who 
was just connected and it would make him very angry ; it sounded 
as if she wanted to hurry him up or check him off. The other 
question, ' Did you get them ? ' seems like a desire to aid him and help 
him, and he answers pleasantly. When a ring-off comes, we believe 
that operators, in nine cases out of ten, will know positively whether 
it is a ring-off or a desire for a second connection, so that we take 
the risk of having them disconnect promptly without asking any 
further questions, and allowing the second connection wanted to 
come in the regular way on the regular subscriber's drop. Our 
experience is that we have made but very few mistakes in cutting 
off parties before they were through, and the advantage gained by 
promptly disconnecting is in having to answer less frequently that 
they are talking. 


It was agreed that the exchange manager should be a man, and 
that he should have some degree of electrical skill. Mr. Sunny 
said their method in Chicago was to promote a boy from the switch- 
board, after he had been at it long enough to deserve promotion, and 
put him in the inspection and repair department. 

He is educated in that department in the work of repairs, and 
we also make him find out how to climb a pole, so as to take out line 
trouble. After he has been in that department two or three years 
and finished the course in the exchange, if he has the requisite 
requirements as to force of character and so on, we make him 
manager . . . We have followed that system for seven or eight 
years, and find that we get the best results. 

Mr. Seely explained that they followed a somewhat similar 
system in New York except as regards pole-climbing. Two of their 
prospective exchange managers tried to climb a pole, ' and we have 
them now as pensioners.' There were eleven exchanges and a 
superintendent of exchanges. ' The superintendent has absolute 
charge of all the managers in each and every exchange, hires all 
the operators, and discharges them, upon recommendation of the 

The operators' school so familiar a feature now was less general 
in those days, but it existed in Chicago and is thus described by 
Mr. Sunny : 

We have a pretty hard time to get girls of the right character 
who are willing to work Sundays, and we are not getting the class 
of applications that we did in the early days of the telephone service. 
This is not because we have always tried to keep the standard of 
intelligence and morality high, but it comes from some other cause, 
of which I know nothing. We probably get one young lady out of 
ten that is entirely satisfactory. We have a school of instruction 
where new-comers are taught the business as far as possible. The 
report of this school runs something like this : For two years we 
have had about three hundred and fifty applicants, and we have 
only found out of the three hundred and fifty about eighty-five who 
were competent to do the acutal work. 

Mr. HALL : That is a very good percentage about one in four. 

Mr. METZGER : What kind of a kindergarten have you there ? 

Mr. SUNNY : We have a teacher. The school will accommodate 
about ten operators ten applicants for positions and we have 
in charge of that school a first-class operator of five or six years' 
experience, who was cut out for a school ma'am. She can size up 
an applicant in short order, and in four or five days can tell with 
almost absolute certainty whether that young lady is going to make 
an operator or not. 

Mr. METZGER : Does she put her down before the regular board ? 


Mr. SUNNY : She sets her down before a little switchboard that 
we have made up for the purpose. We have three sections of 
switchboard, and then telephones scattered about the room, and 
the students talk to each other, and on the three sections of switch- 
board they make connections and do everything they would be 
called upon to do at the big board, except to make the actual 
connections. We find that it educates the students in the matter 
of hearing and talking, and handling the cords and handling the cam 
levers, so that when they sit down to actual work they have nothing 
to overcome except the momentary nervousness. In the old system 
we used to take a new-comer and put her on a section to answer 
fifty subscribers, and we used to depend upon the subscribers .to 
educate the operator and make her competent to fill that position. 

The preference for women operators was already well defined 
by 1887, but difficulties were experienced in the engagement of 
women as night operators. Mr. Smith had overcome the difficulty 
in part, and the need for it he explained as follows : 

I would like to say, with reference to the night operator question, 
that in five of our exchanges we have, within the past year, sub- 
stituted women for men or boys, and with excellent results. In all 
of those places we are called upon to do fire alarm service, and we 
found it necessary to do something to make it more certain, and we 
have accomplished it by having women ; they are wider awake, 
more attentive than young men or boys. We are now considering 
the propriety of substituting women for night service in our Kansas 
City exchange, where we keep thirteen operators for the first half 
of the night. The only difficulty seems to be in getting the young 
women to their homes at that hour of the night. 

It was recognised that wherever the telephone was used at night 
the occasion was one of urgency and as high a class of operators 
as could be found should be selected for the service. 

In operating methods considerable attention was given to the 
relative advantages of having one operator carry out in its entirety 
a required connection or having the work divided, the conclusion 
reached being strongly in favour of having each operator 

perform all the various operations necessary to bring the sub- 
scribers of the lines assigned to her care into actual communication 
with the subscribers which they may call for. This in contra- 
distinction to the mode of central station operation known as the 
' division of labour ' plan, in which the co-operation of two or more 
operators was necessary ; experience having demonstrated that by 
the centralised or single operator method the service is rendered more 
speedy ; the chance of error diminished ; in case of error, imperfect 
service or fault, the responsibility is easily fixed ; and the operator 


becoming familiarised with her subscribers is enabled to more readily 
understand and meet their wishes. 

Operators standing at their work was advocated, one member 
estimating that she could work a third faster standing than sitting. 
High chairs permitting standing or sitting at will, or as occasion 
required, were advocated. 

4. Apparatus. The multiple switchboard had been rapidly 
developed and largely used. There were still, even in the 
conference, advocates of other plans, but the advantages of the 
multiple were so manifest that there was no room for doubt, and 
in the ' Conclusions reached by the Conference/ as narrated in 
Mr. Lockwood's synopsis, the multiple switchboard is given first 
place as follows : 

No. i. The multiple switchboard for large central stations is 
a material improvement upon, and presents decided advantages in, 
the matter of efficiency and economical operation over the grouping 
or sectional switchboard and system, in which a number of lines 
were assigned to a single operator at a given section, and in which 
communications were effected between the several sections by means 
of trunk or transfer lines extending from one switchboard to another 
throughout the system. For such central stations it is decidedly 
to be preferred, and no circumstances can be conceived which 
would render a return to the grouping or trunk line boards desirable. 

The position of the Trunk Board at the beginning of the line 
of sections has the next place. 

No. 2. In central stations where the multiple switchboard is 
employed, that section which is devoted to extra-territorial lines, or 
other out-of-town service, should invariably be the first one reached 
by incoming subscribers' lines ; for when any subscriber's line is 
connected with a distant point through a long line, the connection 
being thus made through the first spring jack, the compound line 
is relieved of any electrical and mechanical difficulties such as 
defective jacks, is rendered less liable to interference, and is made 
to work easier than if the long line connections were indiscriminately 
applied at any point of the course of the subscriber's wire through 
the central office switchboard. 

And the imminence of metallic circuits is recognised in 

No. 3. That the Western Electric Company should hereafter 
counsel its present or prospective customers when considering the 
purchase and installation of large switchboards, or switching 
organisations, to have such switchboards, irrespective of character, 
so arranged as to be readily adapted to metallic circuits. 

The growth in the sizes of switchboards and the large number 


of spring jacks required to be employed were causing some 
apprehension on the score of expense, and the attention of the 
conference was given to the consideration of possible means of 
economy. To place the switchboard horizontally, instead of 
vertically, was regarded even by Mr. Lockwood as a ' more 
hopeful line for future experimentation and consideration than 
any other plan.' Mr. Scribner demonstrated that the operator 
had not the same reach on the horizontal as on the vertical board. 
The conference arrived at no definite conclusion on the subject, 
but it is to be remarked that the exchanges in the United States 
refrained from using flat boards. Some years later the same effort 
at economy was tried in Great Britain and it fell to the writer 
to criticise the proposal, 1 without being aware of the 1887 
conference arguments. The results of the operation of the flat 
boards in Great Britain confirm the criticisms that were made 
against them, and illustrate the advantage of the conference method 
of reaching conclusions. 

One of the principal practical features in the multiple switch- 
boards dealt with by the conference was the cross-talk resulting 
from the test lines. The boards were growing in size beyond the 
expectations at the time of the original design, where the plug con- 
nected the subscriber's line with a test wire, which test wire remained 
connected during conversation, making practically ' two long 
open branches either in one cable with other lines or in two cables 
with other lines ; so there is an electro-static effect between those 
two open branches, and the lines that are talking.' The trouble 
was a serious one, and the conclusion of the conference is thus 
recorded in the synopsis : 

No. 9. The testing arrangement which is now in general use 
in connection with multiple switchboards, whereby an operator at 
any board can determine whether a line wanted is already in use at 
some other switchboard, has been found to be a considerable factor 
in producing cross-talk, because when any two lines are united at 
any switchboard, the test wire of each line is temporarily united 
also to its own line and forms an open branch of the said line. Thus 
there are with each pair of connected lines which are in operation 
two open branches in inductive proximity (that is in the same 
caHf) to similar branches of other lines, some of which are at any 
given moment probably also in operation. Any two connected 
lines are thus inductively united by a condenser with such other 
lines as happen to have their test wires in the same cables and which 
are at the same time being used. 

The result is, that the subscribers of these lines are much dis- 
turbed by overhearing the conversation going on over other lines. 

1 Journal of the Institution of Electrical Engineers, April 16, 1896, xxv. 353. 


This disturbance increases with the size of the switchboard, the 
length of test wire being correspondingly increased. 

To remedy this, a plan has been devised whereby the test wire, 
instead of being grounded through its own main line, can be auto- 
matically and independently grounded, and thus need have no 
electrical connection with its own main line. 

This plan is being experimentally tested, and will be adopted 
and recommended by the Western Electric Company if satisfactory. 

The improvement was outlined to the conference by Mr. Scribner 
as follows : 

It consists in providing a double plug instead of a single plug, 
or with a double cord or sleeve on the plug, which, when the plug is 
inserted, makes contact with the test ring, the sleeve being connected 
with one of the double cords and directly to ground ; so that when 
you connect together two lines, you simply connect those two lines 
one to the other, and ground the test wire directly, which would 
certainly improve the service. 

Although not appearing in the conference proceedings, it may 
be stated here that the new plan under trial was that of Mr. Carty, 
described in his U.S. patent, No. 4I5.765. 1 In giving evidence 
on November 17, 1898, in patent suit against the Detroit Telephone 
Company, Mr. Scribner stated that Mr. Carty explained his invention 
to him some few months before June i, 1885, when the application for 
patent was filed ; that the invention was purchased by the Western 
Electric Company upon his (Mr. Scribner's) recommendation. Mr. 
Scribner in further evidence in the same suit said : 

The test circuit of my patent was a marked step in advance 
over the try circuit of Haskins and Wilson, but it remained for 
John J. Carty to invent the highly developed system of his patent 
No. 415,765, which in detail, as stated in each of the claims, has 
gone into use in practically all the modern multiple switchboards 
made by complainant. 

Admitting the superiority of the multiple system, the conference 
were constrained to inquire what were its limits and what arrange- 
ments would need to be made when those limits were reached. 
Mr. Scribner reported that it was then proposed to equip the 
Cortland Street switchboard with over ten thousand spring jacks 
to each section. That number, however, had not been in opera- 
tion, and Mr. Sargent assumed the then existing limit as five 

Two important features in exchange planning were brought 
out in Mr. Scribner's paper entitled ' The Multiple Switchboard 

1 Dated November 26, 1889 (application filed June i, 1885). 


System, its Requirements and its Limitations.' Hitherto switch- 
boards were installed without definite data as to the requirements 
to be met. Mr. Scribner pointed out that the cost of central office 
switching apparatus depended both upon the number of lines 
and upon 

the aggregate number of connections per day or per hour. The 
aggregate number of connections per day determines the number 
of lines for which one operator can do the work, and hence the 
number of operators, and hence the number of multiple switch- 
boards. Another consideration affecting the number of con- 
nections possible to be made by an operator is the efficiency of 
the operator. There is a great variety in this respect, and there is 
the possibility of increasing the average work by selection and 
by training of the operators. 

The other important new feature was the suggestion to dis- 
tribute busy lines. 

The number of connections an operator may make in an exchange 
may be increased by equalising them among the operators. This 
has never been done thoroughly and systematically. A considerable 
aggregate gain can be made by distributing the lines among the 
operators so as to equalise the work among them. 

The value of systematic distribution of the lines in accordance 
with the number of calls received over them was further emphasised 
during the discussion and bore immediate fruit. A number of dis- 
tributing boards and intermediate distributing boards were sub- 
sequently designed and gradually some form became universally 
used. It is to be feared that outside the influence of the conference 
a considerable time elapsed before the advantages of distribution 
in its effect on the economy in first cost of apparatus and in operating 
expenses were fully realised. The prior practice is well illustrated 
by the remarks of Mr. Sunny : 

I think where a popular mistake is made in determining how 
many wires we ought to put before an operator is in the fact that 
the statistics from the other exchanges are looked up, and we find 
that some exchange in the South is making eighteen hundred 
connections a day by one operator, and that another exchange in 
the North is making eight hundred connections a day by one 
operator, and we immediately split that and conclude that we may 
give to an operator a thousand connections a day, and then we 
will be sure to have a pretty fair service. I think the way to 
determine that would be to find out how the business runs in the 
particular exchange. All over the country it seems to be the busy 
hour, or the busy hours, between nine and eleven. In Chicago the 
busy time is from half-past eight until half-past twelve. Now, 


I would give an operator about as much as she could comfortably 
handle between half-past eight and half-past twelve, and she could 
sit with her arms folded the balance of the day. But when we 
average it up and give an operator eight hundred calls to answer, 
the chances are that you will give her too much for three hours 
a day, and just enough to keep her comfortably busy for the balance 
of the day. 

The capacity of the operator of the period with the best available 
material as well as the principle of distribution is shown in sub- 
sequent remarks as follows : 

Mr. SCRIBNER : Is it not a question of the number of connections 
she can make ? It would seem to me that the whole question is, 
How many connections are you going to have on each line of your 
board during the day, and how many connections can an operator 
make in a day ? or take it for an hour, and the busy hours rather 
than for the whole day, because the connections do not come 
evenly during the day, but you are busier at one hour than at another. 
Take the business for a day, and then take the business for a busy 
hour, and how many connection can one operator make in an hour ? 
Then give her as many lines as will provide those connections. 
Now, we have positive knowledge on the subject of how many 
connections an operator can make in a busy hour. When she is 
rushed we have positive knowledge that she can make two hundred 
and fifty connections. 

Mr. DENVER : Do you mean an average operator ? Would not 
that require quite an expert ? 

Mr. SCRIBNER : It requires a good multiple switchboard operator. 
I do not think it would require an expert. I have myself sat with 
a head telephone upon my head connected in circuit with the head 
telephone of a multiple switchboard operator, and have checked 
off the number of connections she made for an hour ; and I have 
had the thing done by others and counts made at different exchanges, 
and made without the operator's knowledge, so as to prevent any 
special work on the part of the operator ; and these figures vary 
from one hundred and seventy-five to two hundred and twenty-five, 
and in some cases two hundred and fifty connections in an hour by 
an operator. Now, the only thing necessary is to give that operator 
a number of lines which will produce that number of connections, 
and if you give her a set of lines which will produce more than that 
during any given hour, then you break her down. If you accumulate 
in front of her, as is frequently the case, a large number of very busy 
wires, when they become very busy at the busy hours she is broken 
down, and other operators along the line are comparatively idle. 

Mr. DENVER : The only way to give satisfactory service is to 
give prompt answer to the greatest number of calls in the shortest 
possible time ? 

Mr. SCRIBNER : That is the whole secret. By distributing your 


busy wires in front of the different operators, giving each operator 
only what she can attend to easily, two hundred connections in 
the busy hours, which in an eight or ten hour day would amount 
to something like fifteen or eighteen hundred connections. 

Mr. METZGER : Mr. Scribner, do I understand you to say that 
an operator can make two hundred connections per hour and answer 
the subscriber and get the party called for ? 

Mr. SCRIBNER : Get the party called for and get them together. 

Mr. METZGER : Two hundred per hour ? 

Mr. SCRIBNER : Yes, sir ; over two hundred. 

Mr. METZGER : Where is that ? 

Mr. SCRIBNER : I have counted them in New Orleans, and I have 
counted them in Nashville, and I have counted them in Boston, 
which is a slower switchboard than either of these others. I have 
counted them in Boston, where there were from one hundred and 
twenty to one hundred and twenty-five per hour. I have had them 
counted at Omaha in the past week, where a single operator made 
two hundred and twenty-two connections in an hour, and did it 
nicely ; she was not rushed ; it was on a quicker board which they 
have in use there, which system we shall probably have a chance to 
discuss during the week. But the idea is that an operator can take 
care of, without being rushed, from two hundred to two hundred 
and twenty-five connections in an hour ; and if she can do that, she 
can take care of one hundred of the busiest lines that is, one hundred 
of the average lines of the busiest exchange in the country. 

Mr. METZGER : Then, according to that calculation, an operator 
could take care of two hundred wires where the average number of 
connections did not exceed seven per day to a subscriber. 

Mr. SCRIBNER : If they were distributed evenly over the board. 

The facility which the answering jack offered for distributing 
without changing the subscriber's number was not universally 
realised. The following extract from the proceedings will serve 
to illustrate the prior conditions and the need of demonstrating 
even to a selected body the newer practical features : 

Mr. DAVIS : There is one thing that Mr. Scribner has d\\vlt 
on that has not been answered ; perhaps it would be well to answer 
it now ; and that is, the possibility of distributing your busy sub- 
scribers around through the board in such a way as to make the 
work more equal. I do not think in the New York exchange that 
can be done, because each subscriber has his telephone number 
printed upon his paper and on his letter heads, and he objects to 
changing it. I think that alone is sufficient to prevent it being done. 

Mr. METZGER : I would like to state how we do it. We distribute 
among our busy boards the residences, so as to try and relieve 
business on that board, and supposing it to accumulate during th 
year, when we issue a new list we make the changes in that ne 
book, but do not actually make the changes until the book is issu 


Mr. DAVIS : I understand ; but here in New York, the practice 
is for a man to print on his letter heads his telephone number. 

Mr. METZGER : We discourage putting the number on. 

Mr. SUNNY : We have discouraged it for the past year also. We 
ask them to put on their cards the word ' Telephone.' 

Mr. DENVER : We found trouble with it in Boston. 

Mr. SUNNY : We were putting in a section of switchboard in 
the La Salle Street office, and the numbers ran from 2000 up. We 
wanted to abandon those numbers in six or eight months, and we 
told those people not to print those numbers on their cards, that 
we might have to change them. They all fell into line on that. 
I have seen a great deal of stationery with the word ' telephone ' 
and the number omitted. In practice a man pays no attention 
to the number of the telephone printed on a card ; he does not 
hunt up the card to see what the number is, but goes directly to 
the book. 

Mr. SCRIBNER : Mr. Davis assumes that it is necessary to change 
the numbers in order to make this redistribution ; but I do not 
think it is necessary to renumber the subscribers. I think the 
equalisation of the distribution of busy lines among quiet lines may 
be made without renumbering the subscribers. Where you have 
answering jacks and drops, the drops fall and call attention to a 
given answering jack. It does not matter what the number is on that 
drop ; in fact, the number might be left off entirely from the drop 
and from the answering jack, simply putting in place of the sub- 
scriber's number a number to indicate which answering jack to 
go into. In that case the operator would not know the number 
of the subscriber she was answering, would not know his exchange 
number; she would simply know his number upon the board. 
In that way by putting them in their proper position on the face 
of the board for the multiple jacks, and distributing them at will 
through the drops to the answering jacks you would accomplish 
all that you wished to accomplish, and without this general tear- 
ing up of the exchange book and the renumbering of exchange 

Mr. DAVIS : I do not see how that is going to operate. Take 
the reverse thing ; when you call for that person, where is his 
number on the board ? 

Mr. SCRIBNER : On the board you place his number just as it is ; 
on the board he will appear always in his old position ; on the 
multiple jack he would appear in the same position that he now 
holds ; but after it has passed through a series of multiple jacks, 
as subscriber No. I for example, then carry him to drop 23, but 
do not number him there. It does not make any difference to the 
operator what his number is on the multiple jack so long as she 
knows where to go to answer. A drop falls ; she goes into the 
answering jack which corresponds to that drop, and she has the 
subscriber calling ; she does not know who he is to be sure, but she 
has him and can connect him to whomever he wants. She always 


knows him as subscriber No. 23 on her board. That is, you take 
the drop, and answering spring jack and number them from i to 
150, and every subscriber then would have really two numbers 
No. i in the exchange, and No. 23, board 3, in the office ; and, 
connecting with him, it would always be done just as it is now. 

Mr. DAVIS : The operator would have to remember. 

Mr. SCRIBNER : Remember nothing. 

Mr. METZGER : Does not that compel the operator to remember 
the changes ? 

Mr. SCRIBNER : Not at all. The operator plugs into an answer- 
ing spring jack which corresponds to an answering drop. 

Mr. SUNNY : That is all she has to know. 

Mr. SCRIBNER : All she has to do when the drop falls is to know 
where to find the spring jack that goes with that. As they are 
placed near together, and have a definite position with relation to 
each other, they can be numbered the same with relation to each 
other and not with relation to the board. She can go to the answer- 
ing spring jack and find who is wanted. 

Mr. DAVIS : Do you put answering spring jacks on your boards 
now ? 

Mr. SCRIBNER : That is one of Mr. Seely's inventions, and it 
is on either the John Street board or the Thirty-ninth Street ; and 
it takes out that necessity of numbering the drops. 

Mr. SUNNY : They may be lettered A, B, C ? 

Mr. SCRIBNER : They may be lettered A. B, C. 

Mr. SEELY : Upstairs it is so arranged. It is immaterial where 
the call comes from : all the operator does is simply to answer that 

In respect to details of construction there was much discussion 
on spring jacks and indicators. The deleterious effect produced on 
talking through the ordinary indicator was beginning to be pro- 
nounced. The absence of any indicator was one of the advantages 
urged for the Law system, while the special form of clearing-out 
indicator (the ' galvanometer drop ' ) used in the Omaha system 
was recognised as offering less retardation in the talking circuit. 
This was referred to by Mr. Sunny in the following remark : 

Mr. SUNNY : I have an idea that that kind of a disconnecting 
drop would not introduce so much retardation in the telephone 
circuit as the ordinary kind. I don't know what the resistance is, or 
how many convolutions of wire ; but I think that a helix of wire 
with a loose needle in it is not so likely to retard the current as one 
in which the core of iron is close to the wires. 

But this idea was not pursued by the conference. 

In the discussion on spring jacks importance was attached to 
the use of German silver instead of phosphor bronze, when it was 
el cited from Mr. Barton that German silver was first used, but 


that everybody thought phosphor bronze was better. Mr. Barton 
said : 

German silver antedated phosphor bronze as a spring metal in 
electrical instruments, and the first spring jacks had German 
silver springs. Afterwards phosphor bronze came into use, and 
in the large size spring jacks they have been used right along until 
now, and have never given any trouble that I am aware of ; and 
I don't know that any change has been made or has been called 
for in those springs. The people who have room for spring jacks 
of that size have used them right along and preferred them, because 
there is such an abundance of metal there that it makes a contact 
which is very firm indeed. Now, the small spring jacks such as 
are mounted on rubber strips, and such as are used in New York 
and Boston and elsewhere, have a spring of very much less surface, 
and it is a matter of great deal of pains to make them all uniform. 
During the later period of construction of the smaller spring jacks 
the method of testing has been improved, so that now the practice 
is to try every strip of spring jacks by putting a pin into the plug 
hole with a weight at the other end, making a lever of the pin, and 
trying with a telephone at the ear of the man who is making the 
test to see if the contact is lifted. If the contact is impaired by a 
certain weight working against any spring, that spring is marked 
and re jetted and another one is substituted which will not be lifted 
from its contact by that weight. In that way, by taking a great 
deal of pains, the narrow springs of German silver are thought 
to be now quite efficient, and I am in hopes that they will answer 
as well as the wider and stiffer springs of the old type. 

Mr. Lockwood thought all would agree 

that phosphor bronze is not now, and never was for that matter, 
adapted for the purposes of the spring, but such experience as I 
have had with the matter leads me to the belief that this is not 
because it is more easily oxidised than German silver, and not 
because it more easily acquires a dirty surface than German silver, 
but because it did not hold its resiliency. The contact between 
the spring and point is practically permanent ; it is permanent 
except at the moment it is lifted up by the plug when inserted. 
It hardly looks to me possible, if we have a German silver spring 
and a brass or a German silver point, that it will oxidise much, 
because the two surfaces are close together, except at the moment 
when the plug connection is made, and so I think if we have a first 
class German silver spring and any good, clean, unoxidisable metal 
or alloy for pin contact points, we are all right there. Now, the 
question comes in, What shall we do with the plug in the spring 
contact ? It is rather a curious thing it was to me, at least, when 
I first noticed it that I found that the main trouble in the plug 
contact and spring contact was not that it did not make a good 
connection, but that by and by, after the phosphor-bronze spring 


had lost its resiliency, the insertion of the plug failed to lift the 
spring up off the ground contact, left the ground contact there at 
the same time that it made a new contact with another line through 
the plug. 

Suggestions to use platinum in spring jacks and other contacts 
were considered, but Mr. Scribner argued successfully against it. 
His argument was, briefly, that the idea of using platinum is to 
prevent corrosion caused by sparking when a current is broken ; 
there was very little apparatus in telephone work where the current 
was broken or closed, or where sparking and consequent burning 
takes place. It was necessary to consider the possibility of platinum 
contacts in a spring jack causing trouble. 

If we put platinum in a spring jack at all it should be put into 
the contact points and into the spring. It is rather difficult to 
construct a spring jack cheaply in that way. It is rather a difficult 
matter to get an adjustment of parts made as these parts are, by 
machinery, so that the two points would always meet, unless very 
large platinum points be used. If we put the platinum contact 
on one point alone, we will say on the contact point, then the spring 
will be likely to make as much trouble as practically the two points 
make now. Now, in view of the fact that the trouble occasioned 
is not a trouble of corrosion, but a trouble of dust or some foreign 
material collecting between the spring and its contact, isn't it possible 
that there would be just as much trouble in the spring jacks as 
there is now, and is it not possible would be more difficult 
to remove that trouble when it appears than it is now ? The present 
plan is to insert from the front of the board, in case a spring jack is 
found open, a small piece of thin steel not a file, but simply a strip 
of steel, which removes the trouble, which removes the foreign 
material, whatever it may be. It would be next to impossible to 
insert such a thing in case there were platinum points on both spring 
and contact, because you could not strike between the two points. 

Mr. Scribner suggested putting platinum points on the anvil. 
so called, on which the spring rests, but no platinum on the German 
silver spring, thinking that German silver is free enough from 
oxidation to prevent trouble. The improved springs, however, 
were of such a character that platinum was not necessary even on 
the anvil contact. This was fortunate, for subscribers multiplied 
and multiple switchboards grew to dimensions unexpected even 
by the conference. Their cost was considerable, and had platinum 
been used, a very large amount would have been added to such 

The conference devoted attention to apparatus other than 
switchboards. Mr. Sunny presented a paper of which a part relating 
to the magneto bell has been quoted in Chapter xii. p. 149. 


He continued : 

We have on record [in Chicago] 5300 cases of transmitter trouble 
for 1886, and 6300 cases in 1887 (December estimated). This 
includes the transmitter, primary circuit, and battery. Five per 
cent. of these troubles are bad connections in the hinge, which develop 
after the instrument has been in service six months or a year. Of 
course, the majority of the trouble originates in the battery. The 
Leclanche form of porous cup battery is generally used, and the 
current that it furnishes is at its greatest strength when the circuit 
is first closed, and constantly loses every minute it is kept in service, 
so that at the end of a fifteen minute conversation a very perceptible 
difference in the transmission is noted. A transmitter that will 
automatically adjust itself to the changing condition of a battery, 
or a battery with more constancy is needed sorely. We can be 
good-natured through all the switch and battery trouble, but it is 
hard work to be bland and smiling when the telephone itself shows 
defects. Five hundred cases of telephone trouble in 1886 and six 
hundred in 1887 are made up very largely of shavings of rubber 
getting in between the diaphragm and the magnet. This trouble 
began about the time the use of wax was discontinued, and was 
found to exist in a batch fresh from the factory a month ago. The 
aggravating feature about this trouble is that it seems such a 
simple matter to remedy. The total number of cases of trouble 
and inspection for 1886 was 48,406 for 4460 telephones, of which 
thirty-five per cent, was line trouble and sixty-five per cent, instru- 
ment trouble and inspection. For the current year the total number 
will be 49,279, approximately, for 5100 telephones : twenty-nine 
per cent, line troubles and seventy-one per cent, instrument 
trouble and inspections. The cost will average twenty-seven cents 
for each, or about $9500. No man can figure the actual cost to the 
telephone at large of those cases of trouble. They lose us a friend 
here and there ; they frequently determine a would-be subscriber 
that he does not want a telephone ; and they add their no 
inconsiderable mite to what makes up the sum total of the cry of 
' Monopoly.' 

In the discussion which followed, Mr. Lockwood struck a prevalent 
source of trouble, remarking : 

I notice with some interest that a great number of complaints 
arose in practice in reference to the various kinds of telephone 
apparatus from the defective connection of the spring in the hinges. 
It appears to me that there is a point to which both the suppliers 
of the telephones and the manufacturers of the bells might give 
earnest heed. I do not think, if I may express my own opinion, 
that we are doing justice to the apparatus in sticking to the old 
spring, fastened by a screw at one end, and projecting out at the 
other end. I do not think that the connection is a trustworthy 
one ; and it is not for want of knowing better that we stick to it. 


I know at least three persons who are actively engaged in practical 
telephony, who have suggested that a better arrangement might 
be made by allowing the centre portion of the pintle of the hinges 
to be stripped of any connection with the hinges, and then have a 
spiral of stiff wire, brass or perhaps steel, or whatever might be 
found best, which should really be a portion of the circuit, and 
which should be extended out on both sides and soldered to the 
bell or transmitter connections, as it might be. I have a very strong 
impression that that would be a much better plan, and I should 
like very much to see it adopted both by the telephone manufacturers 
and by bell manufacturers. 

This simple remedy was promptly applied to magneto bells, 
transmitters, and other apparatus, thereby reducing a prevalent 
source of trouble, and tending materially to increase the reliability 
of the service. 

Less effective were the criticisms of the design of apparatus. 
Mr. Sunny opened the attack in his paper by remarking that 

The field for improvement in the construction of subscribers' 
apparatus is a particularly broad one. The entire outfit is crude 
and defective, and it represents a smaller growth towards perfection 
than anything else that we have in the service. 

And Mr. Lockwood continued it by saying : 

I think we have also got a valuable idea, at least, in the suggestion 
that we do not make our telephone apparatus as sightly and as 
attractive to the eye as it might be made. 

There were practical objections to changes in design. Mr. 
Seely alluded to the difficulty found in New York in introducing 
a new kind of bell. ' If we put in a new bell for one subscriber 
we will have five or six hundred to put in next week of the same kind 
of bells, and we find it a very expensive luxury.' Other members 
recorded a similar experience, and there is little doubt that the con- 
tinuance of the criticised type, so far as appearances are concerned, 
is largely to be accounted for by the disturbing effect upon existing 
subscribers arising from the introduction of a more attractive type 
to new subscribers. But so far as working parts were concerned, 
improvements were continually introduced as they were called 
for by experience or developed by invention. 

Party lines existed in the earliest days of the telephone, and 
the apparatus in connection with them was the subject of numerous 
inventions mainly with a view to obtaining a system of selective 
signalling which should prevent all the subscribers on a line being 
called when only one was required. By 1887 there was sufficient 


experience of the original form and of its successors to show that 
none was satisfactory. Still the telephone manager was faced with 
the fact that a line was unused for a much greater period than 
it was in use. This seemed an economical heresy, and it was felt 
that the use of the telephone might be extended and the cost 
reduced if some means could be found for the common use of a line 
or lines. Like most other points, the subject was introduced to the 
consideration of the conference on broad grounds. Question 15 
of Mr. Lockwood's paper reads : 

Whether, considering that the lines of subscribers are in many 
exchanges only used in the aggregate a small fraction of each day, 
it is feasible to relieve the central office work, and at the same time 
reduce the cost of construction by some system of rearranging 
main circuits outside ; so that a given number of circuits may 
suffice for a greater number of subscribers, while at the same time 
each subscriber may be enabled to use some wire every time he 
desires ? 

Reference was made to a system suggested by Mr. Berthon of 
the Paris exchange and described by him at the telephone con- 
vention at Philadelphia in I884. 1 It was a duplex or multiplex 
system on the principle of a Wheatstone Bridge. It was an idea 
rather than a practical application, but another method of selective 
signalling for party lines described by Mr. Berthon at the same 
meeting was an early application of a principle that in later years 
was much used. Mr. Berthon described the system as follows : 

One arrangement is for four subscribers on metallic circuit. 
We have four keys at the central office, and when we wish to call one 
subscriber we call him by a positive or negative current sent on one 
or other of the wires, the wires being grounded at subscriber's 
office. That current goes through the ground, positive or negative 
on the wire, and reaches the ground at the other end, so that with 
polarised relays at subscribers we can call the two subscribers 
branched on the first wire and the two branched on the second wire, 
which makes four. This is one of the best combinations we have 
for calling four persons on one circuit. 2 

As remarked, however, the subject was brought before the 
1887 conference not merely as a question of party lines, but with a 
view to the possible suggestion of a ' nucleus or germ of something 
we have not reached yet.' The economic side impressed itself 
upon Mr. Hall, who remarked : 

It has always seemed to me that a system which gave, for instance, 

1 National Telephone Exchange Association Report, 1884, p. 133. 

2 Ibid. p. 136. 


to a residence a private wire and all the equipment of a subscriber's 
and central office stations, all of that apparatus and plant being 
idle for twenty-three hours and fifty minutes out of the twenty-four, 
involved an unreasonable waste. 

This argument was met by Mr. Sargent, who said : 

I have given that idea some thought, and it seems to me that it is 
a question of relative economy. You start out in the first place with 
the proposition that there is a terrible waste of wire and expense ; 
you propose to remedy this by mechanical devices that, up to the 
present time, have been very imperfect, and to make them perfect, 
I think, would require an amount of expense that would bring you 
back to the first ground, and that it is a question as to whether you 
can produce the mechanical effects tor reaching individual sub- 
scribers over one wire or whether you can better afford to multiply 
the wires than to pay for expense and intricate machinery. 

Mr. Barton added some personal experience and, as was the 
case with most of his remarks, struck a definite note of a practical 
kind. Inventors of good ability and the best experience had 
devoted a great deal of attention to, and much money had been 
expended upon, this subject. None of the schemes heretofore 
proposed and tried had been, he said, on the whole satisfactory : 

I had a telephone line built to my house some years ago with the 
expectation and understanding that my next-door neighbour, who 
is an intimate friend of the family, should share in the use of the 
line, and an excellent type of apparatus was provided, which it was 
supposed would reduce the objections to that kind of partnership 
to a minimum ; I am free to say that I would not now, under any 
consideration, be willing to allow any partnership in the use of my 
telephone line. It is not used more than ordinary residence lines 
are used, but when I want that line from either end I want it ; I 
want it then ; I don't want to wait until my neighbour is through ; 
and I presume in Chicago and in other cities of the largest class there 
can be found subscribers enough who are willing to pay the price 
of telephone service including a fair price for individual lines in 
metallic circuits. The question of partnership lines would then 
include not only the items of economy in the apparatus and lines, 
and the maintenance of the apparatus and lines, and the maintenance 
of the service, but also the element of what the subscribers are 
willing to pay on the condition that they have their service without 
any drawback which can possibly be avoided. Waiting for some 
one else to get through is a drawback that can be avoided ; and 
inside of the limits of one hundred dollars a year the exchanges can 
afford to furnish lines without that drawback. 

Mr. Wilson remarked that if a great saving in investment and 
expense can be made by placing two or more instruments on the 


same line, it stood to reason that the telephone companies could 
make a lower rate for that service and that subscribers should be 
given the option of obtaining either direct or party line service, 
according to what they were prepared to pay. 

Mr. Sunny estimated that, after deducting the cost of the special 
apparatus required, the extra cost of operating would wipe out the 
saving in ten years ; and Mr. Sargent added the further element of 
cost of inspection, for from his own experience ' any individual bell 
system is going to be more than ordinary inspection.' 

In the synopsis this point is included in the category of ' con- 
clusions indicated by drift of discussion/ the result being given as 
follows : 

That in view of the obvious fact that in most exchanges the 
subscribers' lines are only used for a small fraction of the -day, and 
that during the remainder of the time they are idle and represent 
unremunerative capital ; it is desirable that some arrangement, 
or system, be devised whereby, for residence lines at least, a given 
number of circuits may be made to serve efficiently a greater number 
of subscribers. 

The last speakers on the subject were Mr. Hall, Mr. Barton, 
and Mr. Lockwood, whose remarks are given below in extenso : 

Mr. HALL : Mr. Wilson brought out one very important point, 
that with the introduction of the metallic circuits and that is 
inevitably coming the margin of profit to the exchange on any 
possible rate is going to be reduced. It seems to me, however, that 
we may properly and rightly hope and expect that invention in the 
telephonic field will keep pace with the demands of the service and 
the exigencies of the situation, so that we can continue to do profit- 
able business and adapt our business to the conditions as they from 
time to time change ; and that is what I had in mind in getting up 
this discussion, that we should get the problem clearly before us, 
and not reject the idea because the methods which had been used 
in the past had not been successful ; and that we should look forward 
to working out this problem in some way in the future to meet 
conditions which we know are going to exist. 

Mr. BARTON : Meanwhile the more imminent problem is to 
adequately and economically serve the subscribers on single lines, 
and then after that has been tolerably successfully worked out, the 
inventors, I hope, can be got to work on the problem which is now 
not so imminent, but which is coming along in the course of time, 
of giving a second grade or third grade of telephone service. The 
first grade, however, is lacking a good deal in many respects of 
efficiency and satisfaction. 

Mr. LOCKWOOD : The argument of Mr. Barton is good, but I 
can recollect a time when we had some switchboards which could be 


worked, but which had a great many troubles in them. While one 
set of inventors and exchange managers were trying to get the 
trouble out of the old switchboard, another set went to work and 
worked up a multiple switchboard ; but if they had waited until 
all the troubles were got out of the old one, it is possible that there 
might not have been such a large field for the new one. I think the 
moral we have to learn from that is, that while it is well to accom- 
plish one branch of the business and do what we have to do now 
and do it as well as we can, we must not wait for the necessities of 
the future, or rather we must not wait for apparatus that will meet 
the necessities of the future, until those necessities are right upon us. 
Mr. HALL : In other words, while we do not want to cross the 
bridge until we come to it, it is advisable now to send the engineers 
ahead to prepare a bridge, if there is not one there. 

These, remarks admirably illustrate the tone of the conference 
held within ten years of the inception of a new industry. There is 
conflict upon the pressing questions of the moment, the possible 
conditions of the future, but unanimity of aim to attain methods and 
systems which should bring the benefits of telephone service to the 
largest number of patrons and at rates adapted to their means 
or requirements. The introductory remarks of Mr. Hall on the 
provision, housing, and care of plant, and the paper by Mr. Scribner 
on the bearings of the traffic on the construction and cost of switch- 
boards are especially noticeable at so early a date. 



IN its earliest days the difficulties of the telephone system were 
those incident to the service itself. The development of new devices 
to accomplish a new service, the application of new methods to 
accommodate the rapid increase in subscribers, kept the pioneers 
busy. But, very soon after the telephone, other electrical services 
came into public use. The civilised world with one accord awoke 
to the possibilities of electricity. Jablochoff, Werdermann, Brush, 
Rapieff, and Lontin developed their arc lamps ; Gramme, Siemens, 
Wallace and Farmer their dynamos, ah 1 anticipating the use of such 
apparatus for what has since their time come to be known as 
' isolated plants.' Edison and Swan followed with their incan- 
descent lamps and the bold intention of a general supply. The 
telephone is probably in part responsible for the early development 
of the central station system of electric lighting, for it was the 
commercial success attending his quadruplex telegraph and his 
carbon telephone transmitter that encouraged financiers to supply 
the means for those experiments on a large scale which enabled 
Edison to demonstrate the possibility of central station electric 
supply. It was this working out and demonstration by Edison of 
the complete system, even more than the details of which the 
system was composed, that entitle him to the recognition which 
he then and has since received. 

The applications of electricity are roughly divided into two 
parts the strong current and the weak current. The telephone 
is included in the latter category. Supersensitive to every 
disturbing influence, the telephone was at the mercy of its stronger 
current rivals. Induction from one telephone circuit to another 
was simply a family trouble partly amenable to discipline, but the 
disturbances arising from extraneous currents were more serious. 
Light was followed by power ; and one of the first applications of 
electric power was fos locomotion. The earth was common to both 
strong and weak, and the claim that the weak came first and were 



entitled to protection did not avail. Lightning was a known e.vil 
which had to be guarded against from the first, but ' strong current 
protectors ' were a later development occasioned by the application 
of electricity to the purposes of light and power. 

In Great Britain the telegraphs were in the hands of the State, 
and some legal powers were obtained to protect the lines from the 
invading currents of stronger rivals. In the United States the 
legislatures held aloof and left telegraph, telephone, and electric 
light interests to settle their affairs amongst themselves with such 
assistance as the existing laws might afford. While the Federal 
authority in these early days did not interfere, the legislatures of a 
number of the States and several of the municipalities did ; their 
interference mainly taking the direction of insisting upon the speedy 
removal of overhead telephone and other electric wires in large 
cities, and of requiring that all such street wires should be placed 
underground. Policies differed, the municipal authorities of New 
York, for instance, backed by a State law, being insistent upon 
immediate subterranean construction and the abolition of pole 
and housetop lines before the art had sufficiently developed, and 
without considering whether this with existent knowledge could 
or could not effectively and satisfactorily be done ; whilst the 
County Council of London, on the other hand, refused underground 
facilities to the telephone company, even when metallic circuit 
apparatus and cabling systems were thoroughly developed. 

The provision of return circuits for the lines was necessary for 
the extension and maintenance of the service as well as for the 
removal of the aerial wires. It was a necessity for the subscribers, 
the operating companies, and the amenities of a city. 

The single wire, or earth return, circuit was general for local 
service, but metallic circuits were adopted for trunk or toll lines. 
The eventual use of metallic circuits for all telephone lines was 
long foreseen, but a period of transition had to be provided for. 
There were single lines and double lines, and intercommunication 
must be possible between them. A test also must be available to 
indicate the engagement of a line. A ' mixed circuit ' switchboard 
had to be devised, and it was necessary to meet this requirement 
with as little change as possible in existing apparatus. British 
patent specification No. 9125, dated July 19, iSgo, 1 describes the 
method adopted. 

The first claim of this patent is : 

In a system of testing to determine whether a subscriber's line 
is in use at another section of switchboard or not, the use of a 

1 Equivalent U.S. specification, No. 442,145, dated December 1890. 


retardation coil, connected from the subscriber's lines in a derived 
circuit, either direct or through a battery to ground. 1 

In this system the spring jacks previously in use were unaltered. 
Metallic circuit lines were connected one limb to the spring, and 
the other to the bush, of the jack. The plug was already double 
as required by the improved testing system of Carty, 2 and the 
connection with the lines was consequently by spring pressure at 
the tip of the plug and by butting contact at the sleeve. Whilst 
this was the first use of such a screen as a retardation coil 
permanently attached between a metallic circuit telephone line 
and disturbing elements, the coil itself was not new, as may be 
gathered from the limiting effect of the introductory words of the 
claim. In prior multiple test systems employing the line as part 
of the test circuit, the earth was there as the necessary condition 
of its use as a telephone line. In the case of metallic circuits in 
which the line was also to serve as a part of the test circuit the earth 
had to be attached, but in some manner not inimical to telephonic 
communication. The retardation coil connected to the cord circuit 
permitted this to be done. The coil as then made was probably 
more effective for the purpose than any of those preceding it, but 
such a coil and its effects were sufficiently well known. For example, 
Dr. Silvanus Thompson described for a different purpose in his 
British patent specification No. 3564 of March 19, i885, 3 an 
' induction plug,' and so called it ' because of the well-known 
property of such an electro-magnet of obstructing, by its great 
electro-magnetic inertia, the passage of rapidly fluctuating currents.' 

A modification of the same principle was adopted by T. D. 
Lockwoodin his U.S. patent No. 393,165, November 20, 1888 (appli- 
cation filed August 10, 1888), in which he describes more in detail the 
electrical characteristics of such coils and uses the expressive phrase 
' it follows that an ordinary electro-magnet is in no inconsiderable 
degree opaque to telephonic or voice currents.' 

The induction plug of Thompson and the ' electro-magnetic 
shunt ' of Lockwood were connected to the line either as shunts 
or in derived circuit (or bridge), whilst the retardation coil of the 
mixed circuit test system was an attachment to a complete metallic 
circuit for the express purpose of adding an earth in the most 
harmless (or least harmful) way. That an earth might be attached 
to a telephone line through a high resistance relay without impeding 
conversation was known in the earliest days. It is shown in 
Scribner's 1879 British patent No. 4903,** but the rapid alternations 

1 British specification, No. 9125, 1890. 2 Chapter xxiv. p. 321. 

3 U.S. equivalent, No. 327,837, October 6, 1885 (application filed April 20, 
1885. Chapter xiv. p. 168. 



of the speaking current and the choking effect of the magnetic 
impedance thereon were not generally understood. Had they been 
so, the clearing-out drop would not have been placed directly in 
the circuit as it was in the standard switchboard x and its successor 
, the multiple. It was not until 1889 that the clearing-out drop was 
made of high resistance, high impedance, and ' teed on ' to the line 
in the same way as the universal switch illustrated in Scribner's 
1879 specification before referred to. 

The multiple switchboard had been developed as a means of 
effecting the required interconnection of the lines with certainty, 
speed, and economy. It underwent numerous changes as a result 
of experience. The overhearing, due to the condenser effect in 
the test line, was not contemplated on its first introduction and 
only became apparent as exchanges grew in size. The difficulty was 
met by the modification of Carty affecting plugs and wiring. Other 
changes of a detail character were numerous, but the principle of 
construction remained the same as when originally introduced 
the line wire was brought to the spring of a jack which made contact 
with a stud, and the circuit was continued from that stud to the. 
spring of the next jack. 2 

The very facilities which the multiple switchboard offered for 
interconnecting the lines tended to the concentration of as many 
lines as possible in one exchange, and this, with the great increase 
in subscribers, resulted in large exchanges. As will have been seen 
in the last chapter, arrangements were in progress in 1887 for a 
switchboard of 10,000 lines, whilst there were actually in work 
numerous exchanges of 5000 lines. Two hundred lines to a section 
were very general, and on this basis a 5ooo-line exchange would 
require twenty-five sections. In every line there would thus be 
twenty-five points of cleavage, twenty-five spring- or hammer- 
contacts, and twenty-five stud- or anvil- contacts. The answering 
(or local section) jack 3 added one more, making twenty-six in each 
line, without considering the trunk line sections. The trunk lines 
being metallic circuit, the jacks required two springs and two studs. 
Assuming such an exchange of 5000 subscribers to require three 
sections of trunk switchboard, there would be six contacts for each 
line, which, added to the twenty-six previously mentioned, would 
make thirty-two vulnerable points within the switchboard for each 
subscriber's line. 

Dirt has been defined as ' matter out of place.' Dust, the 
' fluff ' or ' lint ' from cords or wearing apparel, and other 
material adulterants of the atmosphere are distinctly out of place 
between the hammer and anvil contacts of a spring jack, for, upon 

1 Chapter xiv. p. 173. Fig. 93. Chapter xix. p. 232. 

Chapter xix. p. 238. 


the integrity of the circuit, telephonic speech depends. The 
multiple board, while facilitating enormously the operation of the 
intercommunicating service, increased also this liability to derange- 
ment. It was in a large office that the operating facilities of the 
multiple were most apparent, and it was in the large office that 
' matter out of place ' had the greatest opportunity for causing 
defects. Each jack was in this respect a weak link, and the larger 
the office the greater the number of weak links. 

Spring- jack cleaners small strips of spring steel passed to 
and fro between spring and contact bellows, blowers, and other 
methods of removing 'dust particles were tried. The earliest use of 
the now prevalent vacuum cleaner was probably as a switchboard 
jack cleaner. 

Mr. Wilson related to the October 1891 switchboard committee 
that the Chicago Company had for some time past been blowing 
out the dirt by means of a steam blower, but they found the relief 
only temporary, the displaced dust settled on the cables and found 
its way into the spring jacks again. 

We have consulted the Sturtevant Blower people, and have them 
at work now making an experimental machine which it is intended 
will suck up the dust. The intention is to have a funnel-shaped 
arrangement that we can place right over the face of the spring jacks, 
and the suction of air will be sufficient to draw the dust out of the 
spring jacks. It is also expected that by adapting to it different 
forms of nozzles, we will be able to take up the dust between the 
cables on the back of the board, and, instead of discharging the 
dust in the room, it will be discharged either into a reservoir or 
receptacle for that purpose or out of the windows of the office. 
Just how well the machine will operate remains to be seen. 

The sub-committee on the care of apparatus reported that out of 
a number of schemes which had been proposed for systematic 
cleaning of spring jacks by a draught of air, it seemed better to them 
to adopt that which sucked the air from the face of the spring jacks 
and retained the dust so collected in a receiving chamber or disposed 
of it out of doors, than any scheme which would simply blow the 
dust out of the spring jacks and into other portions of the apparatus 
or into the room. 

The small strip of steel referred to in Mr. Scribner's remarks at 
the 1887 conference, quoted on page 328, was useful, but no real 
remedy existed to overcome the faults due to ' dirty jacks.' 

The defects thus attributable were of various degrees of import- 
ance, from the complete break which could be discovered to the 
increased resistance which only slightly reduced the power of the 
transmitted speech. The faults also were intermittent. One might 

z 2 


exist at one moment and at the next be removed by a current 
from a magneto generator burning out the obstruction. 

The defects in the service caused by imperfect spring- jack 
contacts was one of the most important subjects considered by 
the next switchboard committee held on July 21, 22, and 23, 1891. l 
The proceedings of this committee were not printed, but are recorded 
in type-written volumes, a ' synopsis or abridgment ' preceding 
the full report as in the previous conference. 

From the synopsis I extract the following : 

Consideration of difficulties introduced with the multiple switchboard. 

The present difficulty with the busy test is mainly a concomitant 
of the transition stage of a mixed system of metallic and earth com- 
pleted circuits, and is the result of an attempt to minimise switch- 
board wiring. In such mixed circuits, with the present system of 
testing, false test signals often occur, due to the extensive employ- 
ment of heavy currents in adjacent lines for many purposes, and, 
therefore, the trouble is, or shortly will be, a general one. 

In this connection the sub-committee assigned to this matter 
reported that the existing device is unsatisfactory and why, and 
that a new form of test apparatus said to be capable of working in 
connection with mixed circuits had been submitted, but the com- 
mittee was not yet prepared to report upon its merits. 

The difficulty of maintaining clean spring jacks. The sub- 
committee on the care of apparatus presented its views in a pre- 
liminary way, vigorously stating the necessity of cleanliness in 
person and dress on the part of individuals, and also in all classes 
of apparatus, whether of the exchange station, the central office, 
or the conductor, and emphatically urging neatness in construction, 
thoroughness in repairs, and constant watchfulness. In pursuance 
of this report, which met with general concurrence, it was arranged 
that all members having suggestions regarding the general subject 
should present them to the chairman of the sub-committee. 

In the more specific consideration of the especial difficulty of 
maintaining the efficiency of spring jacks, while the matter was 
referred back to the sub-committee for further consideration, the 
opinion of the general committee was unmistakable, that the only 
radical solution of the difficulty consists in the employment of a 
switchboard which does not involve the use of spring contacts 
in the main circuit. 

In view of the very serious difficulties which occur in practice, 
by reason of unclean spring-jack contacts, this subject is regarded 
as being of the highest consequence, and, as has been stated, a 
sub-committee to take charge of it was appointed. 

1 The members were : E. J. Hall (chairman), T. D. Lockwood, C. E. 
Scribner, J. J. Carty, I. H. Farnham, and A. S. Hibbard. C. H. Wilson 
had also been nominated a member but was unable to be present. 


Prior difficulties had been ingeniously overcome by minor 
alterations, but on this matter the committee were "not prepared 
to consider palliatives. They looked for a ' radical solution of the 
difficulty,' and that involved a new design altogether there must 
be ' no spring contacts in the main circuit.' 

It will be recalled that the break in the jack was required in 
order that the indicator should be inoperative whilst the lines were 
engaged. If the indicators of engaged lines should be operated, 
there would obviously be a condition approaching chaos. The 
cutting out of the subscriber's indicator introduced in the standard 
board was a valuable feature carried on to the multiple board, 
where it was of even greater value in view of the possible distances 
which separated the indicators of the calling and the called sub- 
scriber. To effect it, however, a make and break in the spring 
jack was required. The possible defects from this make-and-break 
contact were of little consequence in the single jack of the standard 
switchboard or the few jacks of the earlier multiple boards. It was 
later, when the larger multiple boards had been for some time in 
operation, when the standard of service also was becoming higher 
and the demands of subscribers more exacting, that the jack troubles 
forced such attention upon them as to prompt the committee to 
express an opinion (which was practically a decision) that there should 
be no make-and-break spring contacts in the jacks. 

Multiple switchboards had been used on the Law system, simple 
socket contacts being employed. The socket sufficed because, on the 
Law system, the attention of the operator was obtained through the 
telephone over a call wire, and since there were no indicators there 
was no need of a device to cut them out of circuit. 

In the July 1891 conference the board on the new type was 
referred to as the socket system. In the attention given to it the 
committee organised rather than initiated the demand. This may 
be seen from the following extract of a letter from Mr. Barton, read 
at the opening of the proceedings : 

Should an experimental switchboard be made on the socket 
system ? If yes, then which of the, say, half a dozen available 
socket systems should be employed ? The socket system leaves 
the ground on the line while subscribers are talking and holds the 
subscriber's drop from falling. After a selection of the most promis- 
ing of the socket systems for further development and study, then 
it will be necessary to take some one central office and work out 
further details of its application to the actual conditions. This will 
take time and undoubtedly require further experimentation and 
involve discussions at subsequent meetings of the committee. 

A plan proposed by Mr. Pickernell was also referred to, but no 


selection was made at this meeting in July. The committee 
resumed in October of the same year (1891), when a sub-committee 
submitted a skeleton specification for a ' branch terminal ' switch- 
board. The name was thus returning to the descriptive form 
which was first used in the 1887 conference. The seventh point in 
Mr. Lockwood's paper at that conference was as follows : 

It has been frequently suggested that in a multiple switchboard 
single branches instead of loops may run to the several sections. 
The advantage of the branch compared with the loop is that spring 
contacts in the line circuit could be dispensed with. Is there 
sufficient promise in the idea to make it worth considering again ? 

And in the discussion Mr. Lockwood said : 

The multiple board operated in the way that the Law- system is 
operated, or in the way that the Philadelphia and St. Louis com- 
panies are operated, is simply the multiple board operated as 
described in this seventh proposition, where, instead of looping to 
each section of the switchboard, the line is branched. 

The multiple with contacts in the jacks was sometimes called 
the ' series ' system, and by comparison the system without contacts 
might have been called the ' parallel.' Mr. Lockwood himself sug- 
gested that name in the course of the discussion in October 1891, 
when he said that in contradistinction to the ordinary or series 
form, that which had been variously denominated a socket board, or 
a parenthesis board, might better be called a parallel switchboard. 
He said : 

From the very beginning of the multiple switchboard use, and 
long before its use, the branching idea has been before the world to 
some extent. In fact, the very first patent ever granted broadly 
covering the multiple switchboard, that to Mr. Firman, was of that 
type. But the difficulty before us now did not come in there, and it 
was possible to leave all the branches normally open or discontinuous 
because the call came in on a separate line ; the call came in by tin- 
American district system. Later on C. C. Haskins and Mr. Wilson 
(here with us now) invented a kind of try signal for just such a switch- 
board ; and sometime subsequent Mr. Sabin of San Francisco got up 
a plan in which the circuit was to be dosed, although a parallel 
switchboard with normally open branches coming into sockets was 
then contemplated ; and the difficulty at once appeared (which we 
have) that there was a line drop in there, that the line drop would 
come down when the second subscriber was called, and that it was a 
short circuit through the normal circuit of the line when the con- 
nection was on that is, when two lines were connected together, 
if it was a grounded circuit, as all were then, the ground would still 
be left on unless some means were taken to get it off. Mr. Sabin 


adopted the means of closing a local circuit, which would trip the 
line drop and which in tripping it would take off the original ground. 
Of course that could be applied, if it could be applied at all, to 
metallic circuits. There are a great many electrical and mechanical 
difficulties in the case. It was not, I think, until a comparatively 
recent date that the present series of devices have been thought of, 
at least thought of systematically, with means to leave the normal 
circuit closed in some way, either through a magnetic resistance or 
through some other resistance, so as to keep the line drop from 
coming down and so as to ensure at the same time that the clearing- 
out drop shall come down "whenever it is wanted. That, therefore, 
is the problem which is at present before us, and which is solved, 
or which appears to be solved, in one or two ways by the diagram 
and description before us. There are other ways which will come 
up later, but I wished to say so much to show the committee what 
the history of the idea has been. 

The term adopted by the 1891 committee was, however, the 
' branch terminal,' and this has continued to be used in the United 
States, but in Europe the system has generally been known under 
the name of ' branching,' which naturally and briefly describes the 
system in which the lines are simply branched into the jacks from 
continuous stems in contradistinction to one in which they are 
looped into and out of the jacks on the several sections. 

The plan proposed in October 1891 included the main features 
of the system which was finally developed, but it included also a 
proposal which was speedily altered. The members of the switch- 
board committee were impressed with the growing cost of multiple 
boards due to the number of jacks, and were anxious to have a 
practical example of a horizontal (or ' flat ' ) switchboard. It was 
realised that the absence of contacts in the jacks materially reduced 
the liability to troubles from dust, and it was thought that a practical 
test might therefore be given to the flat form. The policy of testing 
new plans in a single exchange under sympathetic conditions and 
careful, observation was followed in this case. The exchange 
selected for the branching board was the Tremont Street office 
in Boston in the territory of Mr. Farnham, a member of the com- 
mittee. In October 1891 the committee had recommended the 
adoption of the flat form. At its next meeting, in March 1892, Mr. 
Farnham reported that they had made up a model under working 
conditions, had submitted it to practical tests by operators, and 
given it such consideration as satisfied him that he did not wish 
to have the flat board in Tremont Street. The committee dis- 
cussed the matter exhaustively and decided that it was not expedient 
to install the new board in a horizontal form, though it was advisable 
to continue experimental work in that direction. The first installa- 
tion of the branching system was therefore of the vertical type, 


and the subsequent experiments with the horizontal form only 
demonstrated the more clearly, that any advantages it might appear 
to have in first cost were more than negatived by the inconvenience 
and expense in working. 

The branching system, as finally adopted, was gradually evolved. 
Ideas were submitted to, and discussed by, the committee. Some 
were rejected altogether. Others were regarded as offering promise 
of satisfactory development, and in these improvements were 

There were two prominent exceptions to the general development 
of telephone systems on the earth circuit plan the Telephone 
Company in Paris and the Post Office systems in Great Britain. 

FIG. 112. Double Jack-knife Switch, as used 
at Paris in 1882. 

The latter were small, the principal exchange being at Newcastle-on- 
Tyne. The Paris system was larger. The use of metallic circuits 
was forced upon the Paris Telephone Company by the city authori- 
ties, whose extensive system of sewers provided an already existing 
spacious subway in which to lay cables as an alternative to the sus- 
pension of aerial wires. The cables first used had rubber insulated 
conductors encased in lead and were thus not available for very long 
distances, but the telephone was not then considered a long-distance 
instrument. Conversation was possible within the Paris limits. 

A description of the Paris system appears in ' Bell's Electric 
Speaking Telephone ' byPrescott (1884), this description being taken 
from the Journal Telegraphique, of which the date is not mentioned. 
It was published in the Journal Telegraphique of January 25, 1882 
(vi. 26). Illustrations are given of the ' double jack-knife switch ' 
(fig. 112) and of the ' peg ' used therewith (fig. 113). 

In general form the double jack-knife switch followed the Scribner 



design. 1 It was made of two metallic plates separated from each 
other by a thin piece of ebonite. The anterior plate is perforated 
by two orifices, which extend across the ebonite into the posterior 
plate where they are of a little smaller size. 

The two plates each ' communicate with one of the wires of the 
double line.' 

The pegs, which must be introduced in the orifices O and O', are 
likewise intended for double wires. The part a 
is very much larger than the part b, and these 
two parts of the peg are carefully insulated from 
one another by means of the ebonite handle c. 

The concentric form of double plug was 
very general with electrical appliances, but 
with wear it was quite possible to have one 
contact firm and good and the other imperfect. 
For telephone switchboard work it was early 
abandoned, the development being in the 
direction of contacts made between jack and 
plug at different positions in the length of the 
plug so that the loose fitting inevitable with 
wear was no impairment of any contact between 
plug and jack. 

The indicator of the Paris system was con- 
nected permanently to one limb of the line and 
was in contact with the other limb through the 
spring. The insertion of the plug into the 
orifice on the right hand side of the jack raised 
the spring and thereby disconnected the in- 
dicator. It was customary to disconnect the 
indicator of the calling subscriber and to leave 
in circuit that of the called subscriber by 
inserting the second plug in the left hand 
orifice of the called subscriber's jack. The description says: 

It is proper to remark that the arrangement of the jack-knife 
switch number 15 leaves the indicator of the subscriber corresponding 
in the derived circuit of the two wires over which the communication 
is being held. [In other words, the indicator was ' in bridge.'] 
This arrangement does not weaken the sounds of the telephone, 
and we shall see, further on, that the circuit will even admit of two 
of these derivations without affecting the conversation. 

The indicator was of the then general type and had a resistance 
of 200 ohms. 

FIG. 113. Double 
' Peg,' as used at Paris 


Chapter xix. p. 226. 


The earliest publication of the derived or bridged indicator is 
that in fig. I of Scribner's 1879 British patent, illustrated in fig. 62 
and referred to in Chapters xiv. and xix. This fig. 62 is an outline 
illustration of the Universal switch. Scribner's patent is ' partly a 
communication from abroad ' by George D. Clark, Milo G. Kellogg, 
and George B. Scott. The last mentioned was superintendent of 
the Gold and Stock Telegraph Company, and Mr. T. G. Ellsworth 
was the manager of that company's New York Central Office. 

One of the features of the Scribner patent is the turning of the 
connecting bars to indicate their being in use. This feature is 
covered by a United States patent granted in 1880 to T. Gardner 
Ellsworth. 1 This is probably the portion communicated by Scott 
to Scribner, but the United States specification is strictly limited to 
the revolving bars, no mention being made of the clearing-out 
indicator connected in derivation which is fully described in the 
Scribner specification. 

Prescott says : 

This idea of putting the subscribers' line to ground in the central 
office through a resistance, for the purpose of enabling the subscribers 
to notify the central office when they had finished their conversation, 
originated with Mr. R. G. Brown, then chief operator of the Gold 
and Stock Exchange, and now electrical engineer of the General 
Telephone Company, in Paris, where this ingenious device is still in 
use and proves a valuable auxiliary to the service. 

Mr. Brown first conceived this idea in November 1878. The plan 
of employing spring jacks and other looping devices was not only 
expensive, but, with the kind of switchboards in use then, required 
the employment of two connecting bars one for each line which 
took up valuable space; and, looking about for some substitute 
for the looping-in devices, Mr. Brown ascertained that a telephone, 
or high-resistance electro-magnet, could be attached between a 
telephone line and the ground without perceptibly interfering with 
the transmission of speech, and hence adopted this plan with very 
excellent results. 2 

Haskins and Wilson, in their paper read before the American 
Electrical Society in 1879, say : 

When it is desirable to use but one line for both signalling and 
conversation ... a polarised relay [is] placed in a derived circuit 
between line and ground. . . . This relay is wound to a resistance 
of 300 ohms, and remains permanently connected to the line, thus 
forming a slight escape when connection is made between two 

1 U.S. specification, No. 226,991, dated April 27, 1880, (application filed 
December 30, 1879). * Bell's Electric Speaking Telephone, 1884, p. 233. 



By reason of the extra currents set up by the telephonic impulses, 
the detriment to conversation attributable to this escape is not so 
great as might at first be supposed. Each electrical impulse from 
the induction coil of the telephone transmitter, on reaching the 
magnets of the polarised relay, meets an opposing momentary 
current, originating in the helices, and, as the telephone current is 
nearly instantaneous, this opposition is virtually a continuous 
one. Experiment at the centre of a telephone wire, 
three miles in length, showed that when the relays 
employed were wound to a resistance of 300 ohms, 
no difference was discernible whether its coils were 
inserted in the line or between line and ground 
in a derived circuit. With a higher resistance the 
comparison was in favour of the derived circuit, 
while, with a relay of lower resistance the reverse 
was true. 1 

This use of the relay, as described by Haskins 
and Wilson, was the same as adopted in the Uni- 
versal switch and described by Scribner in his 1879 
British patent. 

It seems probable that the indicator employed 
by Mr. Brown in Paris must have had less resist- 201 

ance and impedance and was therefore less effective 
for the purpose than the relay employed on the 
Universal switch to operate a separate indicator. 

The indicator used in the British Post Office 
system differed from that of any other telephone 
administration. The indicators of both lines were 
directly in the circuit, with results which are thus 

described by Preece and Maier : 


It is found that on lines of any considerable 

length the disturbing influence caused by the self- FIG. 114. 

induction of the indicators seriously impairs the British Post 
efficiency of the circuit, and the plan of working with Office Bridged 
the indicators in ' bridge ' has therefore been Indicators, 
resorted to at several exchanges. The two indi- 
cators are by this means got out of the direct line altogether and 
placed across as shown in fig. [114]. The indicator for tliis system 
is similar to that already described except that the electro-magnet 
is lengthened and wound to a high resistance (1000 ohms) . The coils 
also have an iron casing surrounding each, to increase their electro- 
magnetic inertia. 2 

The bridge method of connection was also described by Preece 

1 Journal of the American Electrical Society, 1880, pp. 51-2. 

2 The Telephone, 1889, p. 223. 


in a paper on ' Long Distance Telephony ' at the Society of 
Telegraph Engineers (I.E.E.) in London on May 13, 1886. He 
remarked that 

while these telephones in bridge are perfectly susceptible to action 
from a steady current, when rapid reversals (which the telephone 
currents are) are sent, then the indicators in these bridges, being 
specially constructed of a large mass of iron high resistance with 
an iron sheath, in fact contain so much self-induction that the 
currents passing through from Newcastle to Stockton have no 
influence whatever on the telephones at the intermediate places. 1 

The increase in resistance from 200 to 1000 ohms, and the addition 
of an iron casing to increase the retardation, made a very much more 
effective bridging indicator than the one described in connection 
with the Paris exchange. 

There is some reason to believe that Oliver Heaviside 
contributed to this important practical development. In his 
' Electro-Magnetic Theory,' i8g3, 2 Mr. Oliver Heaviside claimed for 
his brother (Mr. A. W. Heaviside) the first use of the bridging or 
parallel method. 

Mr. A. W. Heaviside was in charge of the Post Office system at 
Newcastle-on-Tyne, where there were a number of instruments 
upon a single circuit. The deleterious effect of the instruments 
in series soon set a limit to the admissible number of instruments 
on a circuit and to the length of the circuit, especially when under- 
ground wires were included. Mr. A. W. Heaviside found by experi- 
ment that an immense improvement was made by putting the 
instruments across the line as shunts or bridges. This method 
was brought to Oliver Heaviside's knowledge, and he was asked for 
the theoretical explanation. At what period he does not say, but 
he adds that it was found ' the bridges caused a weakening of the 
intensity of the speech received when many bridges were passed/ 
and ' to prevent this weakening becoming inconveniently great the 
intermediate call instruments in bridge were purposely made to 
have considerable resistance and inductance. . . . This tended to 
prevent the currents passing along the line from entering the shunts, 
and especially so as regards the currents of high frequency, and 
allowed them to be transmitted in greater magnitude.' 

These were the features which distinguished the Post Office 
bridge indicators described by Preece and Maier. Newcastle was 
the principal telephonic centre of the Post Office. Mr. A. W. 
Heaviside discussed these problems with his brother Oliver, and 
it is perhaps but a reasonable inference that for this very practical 

1 Journal of the Society of Telegraph Engineers, xv. 289. 

2 Vol. i. p. 434. 


advance we are indebted to an independent worker in abstract 

When writing on the subject, it is probable that Oliver Heaviside 
was not familiar with what had been done previously in the way of 
bridge connections. As we have seen from the Universal switch, 
from Scribner's 1879 patent, and from Prescott's claim for 
R. G. Brown of Paris, apparatus had been connected in derived 
circuit or bridge probably before there were any telephones at 

The clearing-out indicators at Manchester were connected in 
bridge in 1880, and so continued for several years. 1 That electro- 
magnets impeded the passage of telephonic currents was well 
understood in 1879, as may be seen from the remarks on page 6 
(line 50) of the complete British specification of Scribner (No. 4903 
of 1879) : ' The earth connection of the coupling bar being through 
the coils of a magnet in no way interferes with telephonic con- 
versation.' 2 The electro-magnet was ' teed on ' (or in bridge) to 
earth from a single wire. Diagrammatically, to one not familiar 
with telephonic currents, it might appear to be a short circuit, and 
it is perhaps for this reason that the patentee of an early invention 
in a new art explained that such a derived connection to earth does 
not interfere with telephonic conversation because it is through 
the coils of an electro-magnet, clearly indicating the knowledge that 
the electro-magnet offers so great impedance to telephonic currents 
that they will not be diverted by the bridge but will continue in 
the main circuit comparatively unimpaired. 

Whilst this bridging method of connection was old for the 
clearing-out indicator and the principle of its use would render it 
equally available for the connection of numerous instruments on 
a line, there is no evidence that it was so used in the United States. 
The series method was found to be unsatisfactory, and remedies 
were suggested, such as connecting the coils of the electro-magnet 
in parallel, 3 or shunting the coils with a condenser. 4 Both these 
methods were referred to by Mr. Lockwood in i887, 5 but there is 
no reference to the bridging method until the paper entitled ' The 
New Era in Telephony' by Hibbard, Carty, and Pickernell, read before 
the National Telephone Exchange Association in September 1889, 6 

1 The Telephone, Preece and. Maier, p. 350 ; Telephone Handbook, Poole, 
first edition, p. 169. 

2 See also Prescott's reference to effect of magnets in telephones with 
multiple diaphragms, and the application of condensers when many sets of 
apparatus are placed in one circuit (The Speaking Telephone, by George B. 
Prescott, 1878, pp. 23 and 31). 

3 Sargent, National Telephone Exchange Association Report, 1883, p. 52. 

* Ross, ibid. 

6 National Telephone Exchange Association Report, 1887, p. 57. 

Ibid, 1889, p. 39. 


where the method is recommended for the connection of switchboard 
indicators and for the connection of two or more subscribers on the 
same metallic circuit. Preece, in the paper quoted, in 1886, and ' The 
Telephone ' by Preece and Maier, published in 1889, describe the 
Post Office bridging indicator, and Preece, in 1886, as well as Oliver 
Heaviside in his ' Electro-Magnetic Theory ' (1893), said that the 
bridging method was adopted in the Newcastle district for con- 
necting a number of instruments on one line. In the Newcastle 
area underground work was general, and the limitations of number 
or distance would therefore be more quickly reached. The New- 
castle district was consequently a likely locality for the practical 
application of the bridging method, and on the theoretical side the 
benefit to be derived from the association of the brothers Heaviside 
seems obvious. The description of the Post Office bridging indi- 
cator explains for the first time the theoretical advantages of the 
increased electro-magnetic inertia and resistance. 

The iron shield around the indicator coils had also the effect 
of reducing the induction between neighbouring indicators when 
not connected in bridge. It was for the latter purpose that they 
were added to a strip of clearing-out drops by the Northern District 
Telephone Company competing with the Post Office in the New- 
castle and Sunderland district. A supply of such indicators was 
ordered bythat company from the Western Electric Company, whose 
further investigations and experiments demonstrated that the best 
effects were obtained by a single coil completely surrounded by an 
iron casing connected to the core. The result was the production 
of the indicator known in the United States as the ' Warner tubular 
drop ' in contradistinction to the ' Warner drop ' as first applied 
to the standard switchboard. 1 

The accompanying illustrations of the tubular indicator (figs. 115, 
116, 117, and 118) are reproduced from British patent specification 
No. 9571 of 1889.2 

The specification relates that the electro-magnet is formed ' of 
a central core of soft iron, a helix of wire, and a tube also of soft 
iron. The external tube and the central core are magnetically 
connected and together attract the armature, to which is attached 
the pivot arm or catch releasing the indicator shutter. This 

1 These drops were the design of Mr. J. C. Warner, who was born in London 
in 1822 and was employed in making electrical apparatus in the workshop 
in which instruments were made for Cooke and Wheatstone. Subsequently 
he made some apparatus for experimental use by Morse on the line between 
Baltimore and Washington. Mr. Warner entered the service of the Western 
Electric Company before the telephonic era. He installed the multiple 
switchboard in Melbourne in 1885, visiting London on his return. 

* Date of application, June 8, 1889; U.S. equivalent, No. 477,616, dated 
June 21, 1892 (application filed, June 17, 1889). 


arrangement increases the sensitiveness of the indicator and reduces 
the electrical induction between neighbouring indicators.' 1 

No mention is made of its special suitability for connection in 
derivation, but it was immediately adopted as a clearing-out drop, 

FIG. 115. Tubular Indicator (plan). 

being connected between line and earth, this feature of the Universal 
switch of 1878 or 1879 being thus restored in 1890. 

\Yhen, therefore, the switchboard committee assembled in 
1891 with the selection of a branching system as one of its principal 

FIG. 116. Tubular Indicator (section). 

objects, a type of indicator especially suitable for bridging purposes 
was already in practical use. There was no question of the suit- 
ability of the Warner tubular drop for permanent connection in 

1 Date of application, June 8, 1889; U.S. equivalent, No. 477,616, da,ted 
June 21, 1892 (application filed June 17, 1889). 


bridge across the two limbs of the line. But for multiple switch- 
board use two principal problems remained : 

(1) How to prevent the drops falling when two lines were con- 
nected ; and 

(2) How to obtain the required test of an engaged line. 
Hitherto the test system derived its efficacy from connection 

with the line circuit ; in the single circuit system by the direct 
earth of the line, in the mixed or transition system by the earth 
through retardation coil. The masses of wires through a switch- 
board were already somewhat serious, and many telephone engineers 
were reluctant to increase them ; so much so that a proposition 
was considered to simplify the switchboard for metallic circuits by 
using induction coils or translators British specification No. 7465 
(1890) is an example of this. Proposals were submitted in which 
the drops were rendered inoperative in a two-wire system by a 

FIG. 117. Tubular Indicator 

FIG. 1 1 8. Tubular 
Indicator (back). 

margin of current, and the busy test also obtained without an 
additional wire. But the committee did not like ' margins,' and 
had had some experience of the unsatisfactory nature of busy 
tests in contact with earth or line. On these points the committee 
(I quote from the summary) reached these conclusions : 

That comparing the two general plans suggested for securing 
the independence of the ' line ' and ' clearing-out ' drops, i.e. that 
depending on a margin of current, and that depending upon posi- 
tively locking the line drop during a connection, the latter plan is 
to be preferred. 

That a busy test requiring but a single testing contact is desir- 
able, and that for the present a third wire through the switchboard 
for each circuit to secure a thoroughly reliable busy test is 

A double cord connection switchboard is recommended. 

The views of the committee concerning ' Branch Terminal 
Switchboards ' having thus been expressed, the subject was refcrn ! 
to a sub-committee consisting of Messrs. Scribner, Hibbard, and 
Lockwood, which sub-committee, after examining a number of plans 
of securing the independence of the ' clearing out ' and ' call signal 



annunciators, and also of busy test apparatus, decided to recommend 
a plan presented by the Western Electric Company, which combined 
a local and third wire busy test with a positive lock device for the 
line drop, and which also promised to produce an additional feature, 
i.e. the automatic restoration of the call drop. 

The promise was amply fulfilled, and the branching system 
passed by the committee at their meeting in March 1892 was intro- 

FIG. 119. Branching Indicator (plan view from above). 

duced to practical service in a form which subsequently required 
very little modification in essential points. The jack as submitted 
by the manufacturers had a stud for the test. This was objected 
to by members of the committee, who generally preferred a circular 
orifice such as had been used with the series-jack test a wise 
decision, in that it offered a more suitable testing surface and 

f~] * rn 




TWH. > rt< ' 






r 7 " *fc 

FIG. 120. Branching Indicator (plan view from below). 

involved no change to the operators in the method of testing for 
an engaged line. 

The principle of the system is explained in British patent 
specification No. 4428, March 5, I892. 1 

Quoting from the British specification the objects of the 
invention are : 

first, to provide circuits and mechanism whereby the individual 
1 Equivalent U.S. specification, No. 563,250. 


annunciator of a line shall be automatically reset or replaced by 
the operation of making a connection to the line, and whereby it 
shall be rendered unresponsive to signalling currents during said 
connection ; second, to provide an annunciator to respond to the 
signal for disconnection, and means for automatically resetting 
the same ; third, to provide suitable means for testing at any board 
to determine whether a line is already in use or not ; t and, fourth, to 

FIG. 121. Branching Indicator (section). 

avoid all connections and branches common to the different l ; nes 
of the exchange. 

But this specification did not include the final form of jack and 
indicator which were put into commercial use. These will be 
found in British specification No. 17,160, September 26, I892. 1 

Figs. 119 to 122 illustrate the indicator, a detailed description 
of which is not necessary. 

FIG. 122. Branching 
Indicator (rear view). 

FIG. 123. Branching Spring Jack (section). 

Fig. 123 is a sectional view of the spring jack in which the short 
spring is connected with one limb of the circuit, the bush with the 
other limb. The two springs of equal length are the test and 
restoring circuit, normally broken but completed on the insertion 
of a plug. 

Fig. 124 illustrates the plug and its engagement with corre- 
sponding parts of the jack. 

1 Equivalent U.S. specification, No. 533,148. 


Fig. 125 shows the circuits in detail when two subscribers are 

The first switchboard on the branching system was installed, 
as already stated, in the Tremont Street office in Boston. The first 
in Great Britain was at Hull, the next at the Avenue Exchange, 
London. The first on the European continent is believed to have 
been at Christiania, but in that case with a smaller jack of skeleton 

The introduction of the branching system involved no change 
in the method of operation by the subscriber, who continued to turn 
a magneto handle for the purpose of calling the exchange and for 
notifying the completion of conversation. It involved no change 
in the methods of the operators except that it rendered unnecessary 
a part of their previous work, because the calling and clearing 
indicators were automatically restored. 

Apart from the saving of actual work, the operation of making 
a connection was much facilitated by the certainty that the correct 
jack had been selected. On the older system the relationship of 
the jack to the indi- 
cator had to be care- 
fully studied, while 
with the branching 
system the fact that 
the indicator was FlG " I2 4-~ Branching Plug and Corresponding 

restored was a proot 

that the correct jack had been plugged into. The automatic re- 
storation feature was in some quarters considered of such import- 
ance as to give the system the name of the ' self-restoring system ' ; 
but, as has been shown already, the automatic restoration was a 
development not originally contemplated. It was a great improve- 
ment in the nature of a bonus. The main feature was the jack 
without break hence the ' branching system,' the introduction of 
which was advocated at the meeting of the switchboard committee 
in July 1891. The spirit animating the members and the method 
governing their deliberations have been referred to in the preceding 
chapter, so that only the practical result to which their attention 
was mainly directed has been recorded in this chapter. But the 
opening remarks of the chairman, addressed as they were to col- 
leagues equally well informed as to progress and tendencies, 
expressed moreover without any regard to publication or any wider 
audience, afford valuable contemporary evidence of the progress 
between 1887 and 1891. Mr. Hall's remarks apply, of course, 
to the United States companies, but with minor differences 
they will illustrate the progress throughout the world. Mr. 
Hall said : 

2 A 2 



Since the switchboard conference held in 1887 there hasTbeen 
no organised work in the direction of improving and standardising 
our exchange apparatus, although there have been several informal 
conferences held and much good work has been done by companies 
and individuals. The service generally is on a much higher plane 
than it was at that time. New difficulties, however, have since 
presented themselves, and the very impetus which the work of that 
conference gave has resulted in the development of much that needs 
study, revision, improvement and, perhaps, repression. 

In nothing has progress been more marked than in the housing 
of our central offices. Very many of the large offices are now in 
commodious and fire-proof buildings owned by the company, and 
a number of new buildings are being constructed. Metallic circuit 
service is no longer confined to the Long Distance Company, the 
underground problem has lost many of its terrors, and multiple 
switchboards are found in all the offices large enough to use them, 
and in some that are small enough not to. 

To bring the service generally up to the high standard made 
possible by our present knowledge calls for the expenditure of vast 
sums of money, and this committee will undoubtedly do much to 
point out the ways in which progress and economy may go hand 
in hand. 

While our special work is the switchboard, its functions are so 
combined with those of all of the other appliances that it cannot 
well be considered alone, and we must necessarily take account of 
all the elements that go to make up exchange apparatus. 

There has been, I think, a tendency to glorify the switchboard 
beyond its deserts. It is the central and most striking feature of 
the exchange, but it is neither the exchange nor the most important 
part of it. It is the largest single piece of our exchange machinery, 
but the aggregate of the subscriber's apparatus, cable terminals, 
distributing and protecting devices, exceeds the switchboard in 
bulk, importance, and cost in all but the largest exchanges. 

From the engineering standpoint the problem of the switchboard 
for a large exchange is undoubtedly the most important that has 
to be solved ; but I am talking now about the country at large. I 
have in mind some exchanges where most of the money and work 
have gone into the central office, and the service suffering from 
neglect of other matters ; while other exchanges which to-day 
with their old board could give good service are failing to do it 
because the managers have asked for new switchboards and defer 
all improvements until they can get them. 

In my judgment the one great need of our exchanges to-day, and 
one that could be supplied almost without cost, is cleanliness. I 
use the word in a broad sense clean offices, clean lines, clean 
switchboards, clean batteries, clean terminals, and clean operating. 

It seems to me that this committee might add materially to the 
good work which I know it will do in selecting and designing 
apparatus if it could devise some way of impressing on every officer 


and employee of our various companies the fact that no piece of 
electrical machinery ever has been or ever can be devised that will 
work well unless it is kept clean and handled properly. 

The general object for which this committee was appointed is 
to improve our exchange service by selecting what is best from the 
many methods of operation and forms of apparatus in use or sug- 
gested for use. To do this in a way to make the results available 
our plans must be laid out thoroughly and systematically and our 
conclusions embodied in specific recommendations. It seems to me 
that our work naturally divides as follows : 

1. To formulate the problems connected with the switchboard 
and other telephonic apparatus relating to exchange service. 

2. To determine how far these problems are satisfactorily 
solved by existing apparatus and methods. 

3. To pass upon the merits of new devices and methods and 
determine to what extent, if any, they should supplement or super- 
sede existing ones. 

4. To prepare standard specifications for all telephonic 

5. To recommend such action as may be necessary to secure 
the general use of standard apparatus and uniformity in methods. 

6. To suggest such lines of experiment as it may seem advisable 
to follow up. 

The subject is a large one, its subdivisions are numerous, and 
many of them involve technical details that cannot be considered 
to advantage in a general meeting. We can, however, arrange, 
if it be thought best, for sub-committees, either standing or special, 
to which matters requiring prolonged and minute study can be 
referred, and the reports of these sub-committees be passed on at 
our general meetings. 

The introduction of the branching system carried with it the 
employment of the bridging magneto, to which Mr. Carty had 
devoted considerable attention ; and it also involved an improvement 
in distributing board apparatus, thus outlined by Mr. Scribner : 

The switchboard wires are led from a general distributing board 
in the usual manner. Two sets of terminals are, however, provided 
at the inner end of this distributing board instead of one, the 
wires leading to the answering jacks and drops being connected to 
the second set. The two sets are placed adjacent to each other so 
that when a line is to be operated at its regular section the connection 
from its terminals to the answering jack terminals may be readily 
made. When desired, however, the line terminals may at this 
board be connected to any answering jack terminals desired, and 
the work of intermediate distribution be thus accomplished, making 
the use of a second, or intermediate distributing board, unnecessary. 

The branching system switchboard was in itself an important 
production, but it was only one example of the trend of thought of 


the time. When the telephone service was a novelty and some- 
thing of a mystery, some latitude might be allowed in the certainty 
of its operation. But as the use extended and the luxury became 
a necessity, doubts as to whether the service was usable or would 
be free from interruption became unbearable. So far as human 
forethought and care could provide, the service must be reliable. 
The branching system established that reliability within the switch 
board. Careful construction, thorough inspection, and daily tests 
of the outside lines as well as the inside apparatus contributed at 
this period to produce that higher standard of service which recog- 
nised that every subscriber was entitled to talk clearly with any 
other disengaged subscriber without delay and without interruption. 

The power required for locking and restoring the indicators 
was obtained by the use of accumulators, and thus the telephone 
engineer became familiar with stronger electrical currents in his 
own work. In the October 1891 meeting of the switchboard 
committee it was related that accumulators, charged direct from 
the Edison Company mains, were used in Chicago to supply between 
sixty and seventy operators' transmitters. 

These were stepping-stones to further advances which were 
shortly to follow. Telephone engineering was growing with the 
growth in the numbers of stations and also in appreciation of the 
further advantages which might be obtained by reliance on a great 
single source of energy instead of a multitude of scattered batteries 
and generators. 



IT is in the report of the meeting of the switchboard committee in 
May 1892 that we first find reference of a practical kind to ' Common 
Battery Exchange Systems.' The term ' central battery ' has 
very generally been applied to the system, but the original term is 
the more correct, for a battery might be central without being 
common to all the subscribers. Mr. Lockwood referred to one 
such system in his remarks to the committee. This was the plan 
adopted by the London and Globe Telephone Company, who were 
using the original Runnings transmitter. Mr. Lockwood said that 
when the Runnings transmitter was first brought over to the United 
States the inventor specified that it should be used directly in the 
circuit with one cell of Daniell battery for every rnile of line. 
Experiments were made between Boston and Providence. The 
distance is forty-four miles, and forty-four cells were accordingly 
inserted. These were gradually reduced to one when the trans- 
mission was found to be very good. Mr. Lockwood continued : 

In England the Globe Telephone Company worked with nothing 
else than the Runnings transmitter with no induction coil, worked 
on a direct circuit also, and they made a great use also in connection 
with it and they found that was one of the principal advantages 
of the visual signal in the main line to indicate the condition of the 
line. They always had a galvanometer signal in their line when 
the line was working. 

Mr. George Lee Anders was the electrician of the London and 
Globe Telephone Company, a reminder of which company's exis- 
tence is still to be seen in the words ' Telephone Building ' over the 
doorway of No. 31 Queen Victoria Street, London, where their 
offices were situated. 

The use cf the Runnings transmitter without an induction coil 
was to then i ot so much a matter of choice as of compulsion. The 
induction coil, in combination with a transmitter, was protected 



by the Edison patent owned by the United Telephone Company, 
and the Globe Company was consequently compelled to carry on 
its service with direct battery working. The method in use for 
indicating the close of conversation has not been, so far as I can 
discover, a matter of record, but I think the galvanometer plan was 
not in operation on the ' duplicate ' switchboard an early name for 
the multiple supplied to the London and Globe Company. But 
that it was some form of automatic clearing signal is evident from 
the instructions to the subscribers in the first issue of their 'Directory,' 
dated December 1883, in which it is stated ' Hanging up the tele- 
phones indicates that conversation is finished ; therefore do not 
hang up your telephones until you are ready to be disconnected.' 

To explain the use of the plural number, it should be said that 
the Hunnings transmitter supplied was of a portable form. The 
instructions state that ' The transmitter is the instrument with the 
red band.' The hand form was adopted as a means of preventing 
the packing of the carbon. Every time the instrument was used 
a movement of the carbon particles would result. In the Anders 
patent subsequently referred to the patentee says ' the transmitter 
which I employ is one of the type of what is known as Runnings' 
transmitter that is to say, one, the handling of which assures its 

Poole refers 1 to a patent of Anders of 1882 to illustrate the 
difference between ' central ' and ' common ' battery. Anders 
had a separate talking battery between each pair of cords, and this, 
as Poole says, was a central battery, but not a common battery. 

Poole also says that Anders' patent was the earliest suggestion 
of common battery or central energy working ; but, in fact, an earlier 
is to be found. The United States patent No. 243,165, dated June 
21, 1881, was entitled ' Centralising individual batteries of a telephone 
exchange,' and was granted to C. E. Scribner. Fig. 126 is the first 
figure of the patent. In the course of his specification he says : 

Heretofore each subscriber has been provided with one or two 
elements of battery in the local circuit of his transmitter and the 
primary of his induction coil. By the use of my invention many 
or all of the subscribers of a system may use the same battery 
for their primary currents. 2 

He relates the difficulties which hitherto had prevented the use 
of the transmitters of several subscribers in the circuit of a single 
battery, and proceeds : 

I overcome these practical difficulties by means of a Wheat- 

1 The Practical Telephone Handbook, fifth edition, 1912, p. 208. 

2 U.S. specification, No. 243,165, June 21, 1881 (application filed, 
April 4, 1881). 


stone bridge placed at each subscriber's station in the circuit of 
the common battery. In one of the four arms of the bridge I place 
a battery transmitter, which may be of the form known as the 
' Edison transmitter ' or a microphone. The other arms of the 
bridge are balanced to the resistance of the transmitter by inserting 
resistance coils. In the cross wire of the Wheatstone bridge is 
placed the primary of the induction coil, and, in consequence of 
the bridge being balanced, there will be no current passing through 
the cross wire of the bridge and primary of induction coil when 
the system is not in use. In addition to the circuit wire to the 
subscribers, I run to each subscriber an individual wire for talking. 

FIG. 126. Centralising Individual Batteries (Scribner, 1881). 

This individual wire passes through a. switch to the call bell and 
to ground when the telephone receiver is on the hook, or through 
the receiving telephone and secondary of the induction coil when 
the receiving telephone is off from the hook. Speaking into the 
transmitter varies the resistance of that arm of the bridge in which 
the transmitter is inserted. This variation of the resistance of the 
one arm of the bridge causes the main battery current to flow 
through the cross wire of the bridge and through the primary of 
induction coil in vibrations corresponding to the diaphragm of the 
transmitter, and by induction these vibrations are communicated 
to the individual wire, and thence to the distant station to which 
the individual wire may be connected. 1 

1 U.S. specification, No. 243,165, June 21, 1881 (application filed 
April 4, 1881). 


Anders's British patent, No. 749 of 1882, includes a modification 
of the Runnings transmitter, a novel battery, and an exchange 
system. There are two separate systems of battery. One is com- 
mon to the whole of the subscribers for signalling purposes, operating 
the individual indicators to call and to disconnect, the simultaneous 
falling of the individual indicators of two connected lines being the 
clearing signal. The other, or speaking batteries, are central but not 
common. There is a battery in each pair of cords with which 
connections are made. The specification thus describes a complete 
method of centralised battery working. Taking the receiver off 
the hook sends a call ; putting it on again sends a clearing signal. 
The method by which this is accomplished is that known as a 
' fleeting contact,' the receiver switch in its passage from its lower 
to its higher position of rest or vice versa coming in contact with 
an earthed spring. Fleeting contacts have been demonstrated 
to be unsatisfactory in telephonic operations though there was 
no considerable experience at that date, and Anders's patent is of 
interest as an example of centralised energy together with simplicity 
of sub-station apparatus. These attempts of Scribner in 1881 and 
Anders in 1882 to remove the battery from the subscribers' station 
to the central office, whilst of historic interest, cannot be regarded 
as having any direct bearing on the development of common 
battery working as now practised, since the combination of the 
talking current with an automatic signal system could only result 
from extensive and long-continued experience of the requirements 
of telephone service. The circuit wire of Scribner may be regarded 
as economically impracticable, and the system of Anders lacked 
any supervisory signals. 

One of the principal advantages of the common battery system 
as finally developed was the provision of automatic signals at the 
central office controlled by the respective receivers at the sub- 
stations and varied according to whether the receiver of either 
station was on or off its suspension hook. Mention should therefore 
be made here of the system adopted by the British Post Office, 
in which automatic calling signals and a single clearing signal 
(through the simultaneous operation of the two line indicators) 
were obtained, though not with either central or common battery. 
The Gower-Bell instrument was originally used by the Post Office. 
This consisted of a pencil microphone and a large Bell receiver 
fixed in the same case. The sounds were conducted from the 
receiver to the ears of the listener through two flexible tubes. 
When out of use the tubes rested in automatic switches one on 
each side of the case. By removing the tube on the user's left hand 
a signal was operated at the central office. The speaking battery 
was brought into operation by removing the tube on the right. 


When the tubes were again placed upon their respective switches 
the clearing signal was given automatically. The method of 
operation by the subscriber was therefore very similar to that 
subsequently developed in the common battery system, but, 
since local batteries were used both for calling and speaking, 
it was so uneconomical in construction and maintenance as 
not to have been adopted by any other administration. Poole 
says that 

although a very perfect system for a small exchange, its advantages 
are dearly bought, as the frequent attention necessary to the Daniell 
batteries and the constant waste of material constitute very serious 
drawbacks. 1 

Like the branching system in its operating features, the common 
battery system, with all the operating advantages which it afforded, 
was a growth. Its origin was the effort to obtain an economical 
and reliable source of current for working transmitters. The 
Leclanche quite early established itself as the most suitable all- 
round battery for telephone service. Compared with other forms 
it was cleanly and required a minimum of attention. Telephonic 
conversations, in the main, were short and not so frequent but 
that in general the intervals between conversations were of sufficient 
duration to allow for battery recuperation. But with the develop- 
ment of long-distance service the Leclanche was found less suitable. 
In a prolonged conversation the commencement might be satis- 
factory and the conclusion inaudible. 

Blake transmitters were in general use for local service. The 
Blake transmitter was excellent with one cell of battery, but was 
liable to ' boil ' with more. One cell did not suffice for some long- 
distance work. It was consequently found necessary in the United 
States to supply special instruments with carbon-granule trans- 
mitters for;use on long-distance lines. With these sets batteries 
of the Fuller type were generally furnished. The maintenance 
and inspection of such batteries were costly to the companies 
as well as to some extent annoying to subscribers. The ad- 
vantages which would result from getting rid of local batteries 
altogether were emphasised at the July 1891 meeting of the 
switchboard committee, particularly by Messrs. Hibbard and 

Accumulators were considered and dynamos were experimented 
with, as appears from the reports of the committee, October 1891. 

It was at the meeting of this committee held in March 1892 
that the final approval was given to the details of the branching 

1 The Practical Telephone Handbook, second edition, 1895, p. 167. 


system. Two months later, at the meeting on May 20, 1892,* the 
Chairman (Mr. Hall) said : 

We have for consideration this morning the invention of Mr. 
Hayes, the purpose of which is to concentrate all battery at the 
central office. It is not necessary perhaps to say how important 
such an invention would be for the business, because we all realise 
what an enormous part of our operating expense is due to the 
maintenance of batteries at subscribers' stations. If we can get 
rid of that, we get rid practically of 90 per cent, of our inspection 
and a very large proportion of course of all our difficulties, and 
if that enables us also to reduce largely our investment in plant, 
why, it is a saving still further in the right direction. 

Mr. Hayes's paper was entitled ' Common Battery Exchange 
System,' and its purpose is indicated in the opening paragraph : 

This system contemplates the use of a large central battery 
placed at the telephone exchange, which furnishes all the current 
necessary for the operation of all the transmitters whose circuits 
terminate at that exchange. In general design it is similar to the 
common battery system which is now quite generally in use in 
large exchanges for furnishing the current for operators' trans- 
mitters at the switchboard. 2 In that system, however, the induction 
coil is placed near the transmitter, and the telephone receiver is 
in the secondary circuit of the induction coil, as is the common 
custom. In the proposed system, which we have called the 
' Common Battery Exchange System,' the induction coil, instead 
of being near the transmitter, i.e. at the subscriber's station, is 
placed near the battery, at the exchange. The telephone receiver 
is placed in the primary circuit with the transmitter, and is operated 
by the current induced by the secondary into the primary circuit. 
In other words, the ' Common Battery Exchange System ' uses a 
long primary circuit extending from the subscriber's station, the 
secondary circuit being to line, with a receiver in the primary 

The battery proposed to be used consists of eight cells of storage 
battery arranged in series, and as many in multiple as may be 
found necessary to prevent cross-talk, probably five, or in all a 
battery of forty cells, having an electro-motive force of sixteen 
volts. . . . The service given to the several subscribers in an 
exchange will not be absolutely the same, but will vary slightly 
with the distance of the subscriber's station from the exchange. . . . 

1 The members of the Committee were E. J. Hall, T. D. Lockwood, C. E. 
Scribner, I. H. Farnham, J. J. Carty, A. S. Hibbard, F. A. Pickemell, H. V. 
Hayes (present) and C. H. Wilson (absent). 

2 The system referred to was devised by Mr. Carty and is the subject of 
U.S. patent No. 518,392, April 17, 1894, application filed April 14, 1891. 
The transmitters in the Chicago exchange were also operated through storage 
batteries. See Mr. Wilson's remarks in National Telephone Exchange Association 
Report, 1890, p. 88. 


The subscriber's apparatus consists of a transmitter of the solid 
back type, a receiver having a moderately low resistance, the usual 
ringer, a gravity switch, and a condenser. . . . When the telephone 
is taken from the hook the circuit is closed and the condenser is 
cut out, thereby causing the drop at the central office to fall. . . . 

The common battery system seems particularly well suited to 
the long distance work, as by it the expense of battery maintenance 
and inspection can be greatly reduced. . . . 

The clearing out is accomplished by means of a galvanometer 
signal which is positive and sure in its operation and is sufficiently 
compact. . . . The advantages of this system are the decrease 
in cost of maintenance due to the concentration of the batteries, 
the freedom from the necessity of frequent inspection, cheapening 
of subscriber's outfit, uniform excellence of service, greater rapidity 
of operator's service, instantaneous detection of defective local 
circuit, and the fact that a poor local circuit will not unbalance a 
long line circuit. 

After analysing the respective costs of local and centralised 
batteries, Mr. Hayes again referred to the economy in inspection, 
having regard to the fact that the apparatus would be ' reduced 
to the simplest instruments ever put out for telephone work.' 

It would seem that an annual inspection would be all that would 
be required with a consequent reduction of one-fifth to one-tenth 
the expense. 

In the course of this paper it was stated that the common battery 
system was developed first for ' speaking tube ' purposes, an applica- 
tion of the telephone more generally known in Europe as ' Domestic ' 
or ' Industrial ' services. It indicates an installation suited to a 
house, factory or institution, by means of which different depart- 
ments may intercommunicate, but does not provide for connection 
with a public exchange. For the purpose of explaining the genesis 
of the system it will be necessary to interrupt at this point the 
consideration of the Committee's Report, and to indicate from other 
sources the circumstances which led to attention being given to this 
application of common battery for domestic telephones. This 
information is derived from evidence given by Mr. Hayes 1 in a patent 

In 1888 and 1889 Mr. Hayes and his assistants had been giving 
much attention to the development of ' so-called speaking tube 
systems.' On August i, 1889, Mr. A. C. White, one of the assistants 
(whose name is better known in connection with the development 
of the Solid Back transmitter) , made a record of a discussion between 

1 Herzog v. New York Telephone Co., Circuit Court of the United States, 
Southern District of New York, Defendants Record, p. 415. 


Mr. Hayes and himself, and noted suggestions which he felt would 
offer a satisfactory means of accomplishing the results desired. The 
memorandum was accompanied by a sketch (fig. 127), which is 
probably the first diagram of common battery system in the 
direct line of development. 

Telephone 100 Telephone and transmitter. 

^ ^p 100 ohms. 



2600 ohrafl . 

B #2 subscriber. 

FIG. 127. First Sketch r.f Common Battery System. 

In the explanation of the sketch Mr. White says : 

I do not think the discharge from A's line would affect B's, as 
B's line is shunted by '5 ohms the battery resistance. It seems 
to me that such an exchange would be highly practicable. 1 

Mr. Hayes considered the suggestion of great importance, and 
efforts were immediately directed towards a demonstration of its 
feasibility. On September 6, 1889, Mr. Hayes reported to Mr. 
Hudson, the President and General Manager, that Mr. White had 
been experimenting with transmitters used upon the line without 
the aid of an induction coil. Where the circuits were metallic it 
was possible to connect a large number of subscribers to the same 
battery without interference or appreciable loss of volume. The 
advantages to be derived from the simplicity of the subscribers' 
apparatus and the centralised battery were set out and further 
attention recommended. 

The interest of the experimental department in the common 
battery system probably received some renewed impetus in 1891 
through the consideration given to a proposal for a telephone 
installation in the Boston Public Library. This institution was 
intended to surpass any other library in its appointments, and some 
of the trustees discussed with Mr. Hudson the suitability of tele- 
phones for obtaining from the various departments the books 
requisitioned by the readers. Mr. Hudson referred the matter to 
Mr. Hayes, who recommended a common battery ' speaking tube ' 

1 Herzog v. New York Telephone Co., Circuit Court of the United States, 
Southern District of New York, Defendants Record, p. 416. 


system in consequence of the small dimensions of the sub-station 
apparatus. Telephones were not adopted by the library authori- 
ties for the purpose suggested, but the time spent over the considera- 
tion of the matter was not wasted if, as seems probable, the common 
battery system received renewed attention therefrom. 

The next step in the development of the system is recorded 
in a memorandum made by another assistant, Mr. W. L. Richards, 
which is as follows : 


October 15, 1891. 

Mr. Hayes suggests the following arrangement of circuits for 
Long Distance and Local Common Battery Exchange System : 


Also indicator arranged in primary circuit to indicate condition 
of lines : 1 

FIG. 129. 

The results of the work on these lines were reported by Mr. 
Hayes on December 7, 1891, with the sketch fig. 130. 

It is this sketch which forms fig. i of Hayes' United States 
patent (fig. 131), No. 474, 323, which was applied for about a month 
later (January 13, 1892), and granted May 3, 1892. The dotted 
lines in fig. 131 indicate alternative connections for single, instead 
of metallic, circuit lines. 

Returning now to Mr. Hayes' paper read before the switchboard 
committee of March 1892, we learn that 

the Local Common Battery Exchange System has not been developed 
to any extent as the system seemed to be more easily experimentally 
developed in connection with other systems. . . . We have worked 

1 Herzog v. New York Telephone Co., Circuit Court of the United States, 
Southern District of New York, Defendants Record, p. 421. 



out two or three schemes for multiple boards which are more or 
less feasible, but they have not been experimented upon and have 
not been developed in any way. The point that I should like to 



Tr Transmitter. 
T Receiver. 
R Subscriber's Bell. 
C Condenser. 
h Gravity Switch. 

FIG. 130. 

B Storage Battery. 

I Induction Coil. 

G Galvanometer Drop. 

X & Z Subscribers' stations. 

Y Central Office. 

bring out would be the possibilities of improving our service and 
cheapening it by using first the long distance common battery 
exchange system. I think by that, in connection with our long 
lines, we could give a service equal to the service which is given 
to-day and decrease the expense of battery maintenance enormously. 

On such lines the system had been tried on two circuits : to 127 
Purchase Street, Boston, for more than eight months, and on Mr. 
John E. Hudson's circuit at 125 Milk Street, Boston, for about 

FIG. 131. Hayes' Common Battery Patent. 

three months. As a result of the practical tests which were made, 
although of course they were limited in number, it was considered 
that the system was entirely practical and commerically operative. 
It will be observed that Mr. Hayes, whilst not limiting him- 
self regarding the eventual possibilities of the system, adopted a 
thoroughly scientific attitude in its presentation, and suggested its 
use first in connection with long-distance subscribers, the mainten- 
ance of whose local batteries was troublesome and expensive. Much 
work yet remained to be done before the application of the system 
to local and general service. The plans had not been formulated, 
much less worked out. Mr. Hayes in 1892, when his patent was 

2 B 


taken out 1 and his paper read before the switchboard committee, 
had, in the words of Kempster Miller, 

devised the method of current supply to transmitter batteries, 
[which has] come into very extended use in the Bell Companies, 
and it has formed the basis of some of the most successful common 
battery systems in the world. 2 

In 1892 it was a method of current supply to transmitter batteries, 
and ' it was the unanimous sentiment of the committee ' to whom 
it was submitted that the scheme shown 

possesses merit which justified making active efforts to develop it, 
and that those efforts should now take the direction of putting 
it in on working lines under the conditions that would be met with 
in practice. 

The method to be adopted was the subject of much discussion. 
To conduct experiments on Long Distance lines was to submit the 
most expensive outfit to unknown chances, and the Long Distance 
Company would simply be furnishing facilities for trying an experi- 
ment in the interest of the exchanges since the expense of battery 
maintenance was of relatively small importance compared with 
the rental derived from the line. 

Mr. Hall, inferring that the committee did not seem to be very 
strongly of the opinion that they ought to take the chance of putting 
it in on twenty-five working lines, suggested that a twenty-five 
wire board should be installed with a certain number of lines at 
the outset which might be regarded as dummy lines and then work 
one line after another by degrees. Each station was to be treated 
as a purely experimental station, with the understanding, of course, 
that it should be used for the purpose of trial, the regular equipment 
being available to fall back on always in the event of difficulty. 
It was contemplated to pick up by degrees a certain number of 
lines, making a special study of the condition of the wire, resist- 
ance, and other factors. In that way varying conditions would be 
created and studied. 

It was arranged to establish this special service in Boston under 
the immediate supervision of Mr. Hayes and his staff. Thus in 
May 1892 a common battery exchange was authorised and shortly 
after installed. 

In former developments it has been easy to observe the stimulus 
of compulsion. Conditions of the service either as regards the 
growth or the discovered defects prompted remedies. There were 
thoughtful students and some careful studies, but the work of the 

1 U.S. specification, No. 474,323, dated May 3, 1892 (application filed 
January 13, 1892). * American Telephone Practice, 4 th edition, p. 270. 


day was too urgent to permit an undue indulgence in the luxury 
of forecast and the provision of improvements in advance of urgent 
needs. Extraneous sources could produce but little in the complex 
field of exchange service for the very sufficient reason of lack of 
experience. Thus inventive ingenuity in the operating field was 
largely directed to overcoming the defects and meeting the require- 
ments of the then present or the immediate future. The telephone 
was a remarkable invention in more ways than one. Its production 
was timely, but so far from its filling a want which had been keenly 
felt, as has been shown in earlier chapters, the public were sceptical 
of its utility and by no means ready to take early advantage of the 
facilities it offered. The same may be said of the first exchange 
systems and of accessory apparatus. But once the advantages 
were apparent and the service fairly launched, it was as true in the 
nineteenth century as Farquhar expressed it in the seventeenth 
century that necessity was the mother of invention. It was 
necessary to develop methods to provide for ever-increasing demands, 
and it was necessary to overcome the defects which the very growth 
produced. Skilled men were limited in number, and they were 
kept very busy in supplying the needs of the moment. 

There were keen students of individual problems, such as Van 
Rysselberg in Belgium, and men who were adopting a scientific 
attitude towards the exchange system as a whole, like Wietlisbach 
in Switzerland ; but only in the United States were there organised 
efforts at advances and improvements in the Exchange methods. 

By the year 1892, when the common battery system was 
presented, the companies had ceased to be surprised at increasing 
demands for service. They had acquired experience, and the staffs 
of capable experts had so far grown in numbers as to permit the 
leading managers and engineers to devote some time to the con- 
sideration of probable future requirements, to look to the possibly 
desirable, and not to limit their attention to the absolutely necessary. 

One direction to which, about this period, considerable attention 
was given was the reconsideration of the multiple switchboard. 
Some practical and strictly economic minds were stih 1 exercised 
at the multiplication of connecting points, some of which might 
never be used, and advocates were found for a divided board 
system. It was the province of the committee to adjudicate on 
this question ; to determine whether some of the money spent upon 
a multiple might not be saved by the division of the work between 
two or more operators. The inquiry took two forms. One was 
an actual analysis of the work on a multiple board, the other the 
preparation of definite proposals for a divided board. 

It was well recognised that for celerity and certainty the one- 
operator method had a great advantage, but with the growth in 


the number of exchanges there were already a large number of calls 
which of necessity could not be completed by one operator. It was 
considered that whilst a divided board might be slower, the effect 
of its use would be to obtain a consistent average of service. 

Fig. 132 is [a facsimile of an outline diagram to illustrate 
Mr. Hall's paper submitted to the committee held in May 1892. 

EE represents a number of sections of connecting board, and 
FF a row of terminal boards. 

A and B are subscribers' stations. A's line goes through a 
spring jack at G on the connecting board, extending from there 
to the spring jack H at the terminal board R, and then to the 

FIG. 132. E. J. Hall's Divided Switchboard. 

terminal drop C. B's line goes through a spring jack at J on the 
connecting board to the spring jack and terminal drop at O. A's 
call is received by the operator at the terminal board R, who inserts 
answering plug in spring jack H. This engages A's line, and since 
there is also a means of access to A's line at the connecting board, 
it becomes necessary to advise the connecting board operator of 
such engagement. Thus, when a plug is inserted in A's line at H, a 
visual signal R is automatically displayed at the connecting board. 
The call of A for line B having been received by the terminal 
operator at R she communicates with J over an order wire. The 
operator at J connects line B to an office trunk extending to R. 
The insertion of the office trunk plug into B's jack at J disconnects 
the end of the line extending from J to O and actuates the visual 
signal M, indicating that station B is busy. In the event of B 


being busy at the time of A's call, the operator at J would take 
the plug representing the end of the office trunk and insert it in 
a special spring jack which automatically causes the visual signal 
K to flutter, and provides also an audible indication of engagement. 
It is not intended to describe in detail the operation of this proposed 
board. Many complex diagrams and elaborate descriptions were 
submitted to the committee and formed the subject of analysis, 
comparison, and debate ; but the above simple diagram and brief 
description comprise the essential principle of the divided board 
discussion to the line of progress. 

In addition to the divided board above described, a system with 
a very similar title but on an entirely different principle was 
developed by M. G. Kellogg. This was called the ' Divided Multiple 
System,' and contemplated the division of subscribers into various 
multiple groups with a selective system of calling by the subscriber. 
Numerous patents were taken out on this system, but for a descrip- 
tion of, and discussion on, the principal features reference may be 
made to the Journal of the Institution of Electrical Engineers, 
xxxii. 795. This system did not meet with the approval of 
telephone engineers generally, but was installed on a large scale 
by an ' Independent ' Company at St. Louis, and though in 
operation for a considerable period was eventually discarded. 

Other attempts to avoid the use of multiples were the express 
systems of Sabin and Hampton in San Francisco and Hibbard 
in Chicago. In the Sabin system ' small sections of switchboard 
were placed contiguously , each section being arranged for three 
operators. The two end operators were answering operators, and an 
intermediate operator served as a trunk operator. The answering 
operators answered calls from subscribers whose lines were located 
at the section at which they were seated and the trunk operators 
completed calls from other sections to all subscribers at this section.' l 

Order wires and transfer circuits available to the various 
operators provided for the interconnection of the various lines. 
Thus an answering operator requiring a connection at the board of a 
trunk operator, by depressing the key leading to that operator, would 
give the order for the subscriber wanted, and the trunk operator 
would designate the transfer line to be used for the purpose of con- 
nection. Such methods were obviously cumbrous compared with 
the multiple, but the delay incident to connecting and disconnecting 
was sought to be minimised by the use of signals automatic in their 
operation and conveying information to the respective operators 
employed in a connection. 

In the Hibbard express system there were some modifications 

* k Scribner, Herzog Case, p. 548. 


of the Sabin system, but in its purpose and in the use of signals 
it was very similar. Whilst adopted for purposes of economy, it 
appeared to one who inspected it in 1895 

to be a very expensive system, seeing that when I was in the 
exchange nine operators were being employed to attend to rather 
less than 300 subscribers. . . . The board is, however, only an 
experimental one, and no doubt will provide valuable data for future 
use. I do not understand the term ' express ' being applied to it, as 
it must, I am sure, be somewhat slower than the multiple. 1 

The chief interest in these express systems is to be found in 
the adoption of the method of calling that has been more particularly 
identified with the common battery system. The subscriber's set 
consisted of transmitter, telephone, ringer, and condenser. 

In Chicago ' The subscriber calls by lifting his telephone from 
the hook, so causing a lamp to light in front of his operator.' 1 

In San Francisco the operation was the same, but the signal 
was electro-magnetic. 

The divided board and the express systems had for their object 
the attainment of economy by the elimination of the multiple. 

The multiple board was not deposed, but the feature that was 
relied upon to reduce the drawbacks attendant upon the employ- 
ment of more than one operator for a call was the use of a system 
of visual signals which should give the respective operators informa- 
tion without the need of speech. Whilst it was found that these 
signals did not make the divided board a practical or economical 
device in one building where a multiple could be installed, there 
would be obvious advantage to their use in the transfer of calls from 
one exchange to another, where the employment of two operators to 
one call was not a question of choice but one of necessity. 

These visual signals were introduced for trunk line working 
between exchanges. Mr. C. J. Phillips in the report quoted above 
described the system as in use in New York, and said that 

owing to the use of automatic signals much less talking is done on 
their call wires than on ours, consequently operators do not have 
to wait to get their calls in, and the listening operator is able to 
attend to a larger number of junctions (30 as against 24) . l 

In the original application of these automatic signals is to be 
seen an effort to provide contrivances which should make practicable 
a switching mechanism which was comparatively imperfect. But 
the evidence which demonstrated the superiority of the multiple 
also demonstrated the great utility of automatic signals, and 
telephone engineers who were anxious to obtain the most perfect 

1 C. J. Phillips, Report to National Telephone Company in London, 1895. 


apparatus with which the public could be served were not slow to 
combine the best in all respects. In his patent of 1892 Mr. Hayes 
had provided automatic signals for calling and clearing ; of the 
latter one only was shown in the diagram (fig. 131), but the pro- 
vision of two, and their automatic operation, were contemplated in 
the text. 

At the meeting in January 1893, Mr. Durant described to the 
switchboard committee l improvements which had been introduced 
in the Law system as carried out at St. Louis. These concerned 
the transfer of calls from one exchange to another, and since, in the 
Law system, the operators who attend to receiving calls cannot for a 
moment leave the work of receiving orders, the signals were required 
to give varied information. The signals proposed were incandescent 
lamps, and the code contemplated their use in three conditions, dark, 
bright, and an intermediate stage which may be called dim. The 
system as presented did not meet with the approval of the committee, 
but (quoting from the summary) 

incidentally the question was raised whether a system of glow 
lamp signals, like that submitted, is a perfectly safe adjunct to a 
central station, and the committee stated its opinion that it is. 

The committee had incidentally reached an important conclusion 
without, perhaps, realising how near was the time when it might 
be put in operation. In the diagram of the divided board (fig. 132) 
the visual signals are apparently intended to be lamps, but in the 
discussion the particular form of signal was not defined. 

In the summary of the January 1893 meeting it is recorded that 

at the request of the chairman Mr. Hayes reported progress in 
the centralised transmitter battery exchange system, and stated 
his conviction, that in a short time as good service could be given 
by it on the longest lines as could be given by the plans and 
apparatus now in use. ... It was thought that by the next 
meeting, results could be reported in such shape as to afford a basis 
for profitable and instructive discussion. 

It was probably not expected that a period exceeding two years 
would elapse before the next meeting of the committee, and in one 
particular it is to be regretted, since we are the less able to trace 
from carefully recorded sources the various stages of progress during 
an important period of experiment and development. 

It was in May 1895, at the office of the American Bell Telephone 
Company, 125 Milk Street, Boston, that the committee next met. 2 

1 Members : E. J. Hall, T. D. Lockwood, C. E. Scribner, I. H. Farnham, 
J. J. Carty, A. S. Hibbard, and F. E. Pickernell. 

z The members present were : J. F. Davis (chairman), E. J. Hall, T. D. 
Lockwood, C. E. Scribner, I. H. Farnham, J. J. Carty, A. S. Hibbard, 
F. E. Pickernell, H. V. Hayes, and W. S. Ford. 


The advance which had been made is indicated in the notice con- 
vening the meeting, which stated that 

Since the last meeting of the switchboard committee held in 
January 1893, there have been proposed, and to a considerable 
extent put into service, new methods or new combinations of old 
methods and new apparatus for operating a telephone exchange 
which have attracted very general attention, and have been now 
so far developed and tested that it is thought desirable that a 
meeting of the committee should be held at an early day to consider 
that which has been done and to mark out, as definitely as practic- 
able, the best path to follow in further development ; also to decide 
what system of service and style of apparatus it is most judicious 
for telephone companies in immediate need of new switchboards 
to order. 

The drift of thought is further indicated in the following : 

You are also requested to collect and bring with you, or, if unable 
to be at the meeting, to send any data within your reach bearing 
upon the design and use of automatic signals, or other improvements 
in equipment of central offices or subscribers' stations, and upon 
methods of operating. It appears to me desirable that the 
committee should give an opinion upon the following matters, 
together with such others as the committee may select. 

ist. Is it recommended that on switchboards hereafter purchased 
provision shall be made for automatic signals in line and connecting 
circuits, which shall inform the operator of whether or not the 
subscriber's telephone is hung on the hook ? 

2nd. If yes, what system of automatic signals, in the opinion 
of the committee, is most promising ? 

3rd. Is it recommended that on switchboards hereafter 
purchased, provision be made for supplying current to sub-station 
transmitters in some other manner than by means of primary 
batteries at the sub-stations ? 

4th. If yes, what method of supplying current to transmitters 
is, in the opinion of the committee, most promising ? 

7th. The best method of counting calls in central offices, for 
statistical purposes, and manner of tabulating and analysing such 

The committee was unanimously of opinion that in future 
switchboards provision should be made for automatic signals in 
line and connecting circuits which would inform the operator 
whether or not the subscriber's telephone is hung on the hook. 
It was further agreed that incandescing or glow lamps should be 
the preferred form of signal device, and that these should be operated 
direct, if possible, but with relays to such an extent as might be 


After discussing the subject fully the committee voted an 
emphatic recommendation that provision should be made for 
supplying current to sub-station transmitters in some other manner 
than by means of primary batteries at the sub-stations. 

It is, however, necessary at some time to come from the general 
to the particular, and this the committee forthwith proceeded to 
do. Three plans were before them, thus related in the synopsis : 

The ordinary centralised and common transmitter supply plan 
was described by Mr. Hayes ; a common battery system proposed 
by Mr. Dean of St. Louis, by Mr. Durant ; plans in which the 
common source at the central station is supplemented by a Plante 
Cell at the sub-station by Messrs. Hayes and Scribner ; and one 
in which a storage battery at the sub-station is charged from the 
central station source while the line is at rest, and discharged to 
furnish transmitter current in operation, by Mr. Hibbard. 

For convenience these plans were, during discussion, referred 
to respectively as the ' Hayes Common Battery System,' and 
the 'Dean/ 'Plante/ and 'Local Storage' Systems, plans or 

For further convenience we may reduce these to the first and 
the last. 

The Hayes common battery system, as previously related, 
had been under test upon a few long distance subscribers' lines 
in Boston. It was reported to the 1895 committee by Mr. Spencer 
that it was first put in operation in Lexington in December 1893, and 
had continued in operation with satisfactory results. It was next 
put in at Wellesley on much longer lines and under much more 
trying conditions ; and about sixty stations in Philadelphia had 
been in operation for about one week. 

The local storage system was a method which had been used 
in a section of the Chicago exchange. Instead of a primary 
battery there was a small accumulator at the sub-station. The 
accumulator was charged during the idle time of the line and pro- 
vided current for the transmitter when the line was in use. About 
100 cells had been in use for three months with satisfactory results. 

The competition between common battery and local storage 
was decided in favour of the latter as recorded in the synopsis 
as follows : 

The opinion of the majority of the committee (Messrs. Carty, 
Hayes, and Ford dissenting and preferring the common battery 
system) was that the ' local storage ' system was the most promising 
plan which had been discussed and that it should be provided in 
connection with switchboards for the immediate future. 

In view of subsequent developments, it is of interest to refer 


more in detail to the opinions expressed by Mr. Carty, who (of 
the minority) occupied an entirely independent position so far as 
the development of the system was concerned. He said : 

The general result of our experience with that system leads 
me to believe that, providing the question of transmission is settled 
satisfactorily, there is nothing that we can demand of the system in 
the way of adapting itself to switchboards that we cannot get from 
it. I think that the general features of the common battery system, 
that we are now discussing, lend themselves beautifully to what 
seems to be the improved trunking and connecting systems. 

The saving clause on the question of transmission was due to 
the fact that the tests in New York were not ' up to the mark,' 
and this was believed to result from the use of an unsuitable type 
of repeating coil, which could be remedied. Mr. Carty continued : 

Now I feel that if we were satisfied that the transmission was 
all right, we would go ahead in the Harlem central office, on which 
we are delaying work, waiting to find out the developments of this 
case. We would go ahead and instal, very likely, a switchboard, 
employing the common battery system of the Hayes type, unless 
something in this meeting should transpire to show that there was 
something better. ...... 

The plant in New York was intended to be the best expression 
which we could give of the Hayes common battery system, and we 
followed as closely as it was possible the latest ideas in that system 
up to the time of its installation. The results were somewhat 
uncertain as to the talking, and the plant was investigated in the 
ordinary way, by competent people, but those who had not special 
experience in the common battery system ; and, in order to be sure 
whether we had what Mr. Hayes would call a correct expression 
of his ideas, I asked one of his assistants while in New York, with 
Mr. Hayes's consent, to make an examination. I am awaiting his 
report, and I will try to see him while here. 

On the following day the local storage system was under 
discussion and the members were asked individually to express 
their opinions. I give Mr. Carty's remarks in extenso : 

Mr. CARTY : My objections to what is known as the Hayes 
common battery system have been removed, with the exception 
of the objection that I raised to the transmission. It has not been 
clearly established that the transmission is all that was claimed for 
it by Mr. Hayes, but my experience with it, and the reports that 
we have had at this meeting, have simply served to cast doubt 
upon the curves which were shown to us. With the single exception, 
then, of the doubt about the character of the transmission, I would 
favour the Hayes common battery system, and favour it because of 


its beautiful simplicity, by the fact that the subscriber's station 
apparatus is simple, especially when you consider such apparatus as 
being extended to a subscriber's desk. 

The absence of the battery and the wire that is needed for it 
is a very good point, when you consider it from the standpoint of 
a subscriber's station. Now, I am informed by Mr. Hibbard that 
the switchboard arrangement which would be suitable for the local 
storage system such as he advocates, would also be suitable for the 
Hayes common battery system. If that is the case and I had to 
put in a switchboard to-morrow, with no further information than 
that which is now before the meeting, I should arrange for a switch- 
board with the Hayes common battery system, and I believe that 
if it is not exactly right now it can be made right ; but, if I was 
mistaken in that latter supposition that it could be made right 
ultimately why, we could then put in our local storage system at 
the subscriber's station. That is all. 

The decision of the majority of the committee was reached with due 
regard to the paramount importance of transmission and on the basis 
of the evidence of practical results then available. The common 
battery system, as then submitted, was but a sketch compared to 
the finished picture it later became. The more credit, therefore, 
seems due to the minority and particularly to Mr. Carty for his 
perception of what it had accomplished and might be made to 

The ' beautiful simplicity ' of the system, the fact that ' there 
is nothing we can demand of the system in the way of adapting 
itself to switchboards that we cannot get from it,' were observations 
on established facts facts important enough not to give a verdict 
on the basis of other established facts that the transmission was 
not in all cases satisfactory. There were reasons for thinking that 
the unsatisfactory features were not incident to the system but 
could be removed, and the faith found expression in the statement 
' I believe that if it is not exactly right now it can be made right.' 

The belief was justified, but how remains to be told. 

The diagram of cord circuit in fig. 133 was presented to the 
switchboard committee of 1895 : 

Here, it will be seen, the repeating coil assumes the form finally 
adopted, the battery being connected to the middle of the coil 
instead of one end as in the preceding diagrams (figs. 130, 131). The 
description is as follows : 

The cord circuit consists of two two-conductor cords connected 
to a split induction coil ; at the centre of the induction coil is a 
battery ; a clearing-out signal is placed in one wire of each cord. . . . 
The visual signal in the cord circuit operates so long as the subscriber 
is connected to the circuit. As soon as his telephone is placed 


upon the hook, the signal ceases to operate, indicating to his operator 
that disconnection is needed. 

Fig. 134 is the drawing which illustrates the first of the modern 
type common battery patents taken out in Great Britain, 1 and 
may probably be regarded as describing an operative system rather 
than as indicating the instrumentalities to be employed. 

The form of subscribers' station apparatus shown is that which 
was suggested in Hayes's 1891 sketch and 1892 patent a call bell 
and condenser though a high-resistance bell without condenser 
is mentioned as an alternative. Lamps and relays were doubtless 
at this time under consideration but had probably not reached a 

d H 


FIG. 133. 

G No. 15 Induction Coil. 
H Common talking battery. 
I Clearing-out signals. 

L Answering plug. 
M Clearing plug. 

stage that could be regarded as reliable. The common battery 
system is therefore illustrated with devices which were known to 
be operative but had only been used on a small or experimental 
scale. The line indicator is of the branching type, the clearing- 
out indicators or ' visual signals V V" ' are not described, but 
indicators of a galvanometer type had been used by Mr. Hayes for 
the purpose of giving the supervisory signals. This diagram 
probably represents the stage which the common battery system 
had reached early in 1895. There were two features ' a common 
battery or centralised source of electric current ' by means of which 
call signals could be sent ; and, secondly ' a common battery or 
centralised source of electric current by means of which telephonic 
transmitters at the subscribers' stations are energised and by means 
of which various automatic signals or clearing-out drops are operated 
in the central station at which the subscribers' lines terminate.' At 
this time some importance was attached to the further statement : 

1 British specification, No. 8222, April 25, 1895. 



' These two features can be used either together or separately.' 
There was, as we have seen, some fear as to transmission, and in 
the first practical application on a large scale at Worcester (Mass.) 
one feature only (the signalling) was used, the transmitters being 
energised by local batteries. 

The imperfect transmission, which caused the majority of the 
switchboard committee to vote against the common battery system, 
was found upon further investigation and experiment to be due 
to the first form of supervisory relay and the connecting cords. 1 

In June 1895 another British patent was applied for (No. 
11,549), which includes a number of methods of putting the common 
battery system into operation and includes some of the devices 
which were finally adopted, including the relay and lamp. 

Allusion has been made already to the use of incandescent lamps 
at St. Louis as described to the 1893 switchboard committee. But 

tvAi'cA, ttterAs fArocyA. 
and ftrevfnfs re fay 

FIG. 135. O'Connell Lamp Signal described by Wilson, 1891. 

an earlier use, designed by O'Connell, was described by Mr. Wilson 
to the committee in 1891 in connection with a trunk line signalling 

Several forms of target indicators have been designed and 
given a good degree of satisfaction, but the most satisfactory results 
are obtained by an illuminated indicator of simple construction, 
designed by Mr. O'Connell. 2 It consists of a hollow metal tube of 
one inch diameter, closed at the outer end, but through which end 

1 Carty, Herzog Case. 

* ' J. J. O'Connell . . . suggested in 1888 the employment of an incan- 
descent lamp upon burglar-alarm circuits in order to permit the legitimate 
occupant of a protected room to send to the alarm office an identification 
signal whereby his entrance to the premises could be made known ' (' The 
Evolution of the Line Signal,' by A. V. Abbott, Transactions American 
Institute of Electrical Engineers, June 29, 1898, xv. 430). 



is cut, stencil fashion, the number of the indicator. Within the 
tube, which is about two inches in length, is placed a miniature 
incandescent electric lamp operated in the local circuit of a relay. 

Fig. 135 represents the lamp portion of the diagram accom- 
panying Mr. Wilson's paper. 

The lamps used at St. Louis in 1893 were also for trunk line 
signals, and were probably, like those of O'Connell, of comparatively 
large dimensions. 

It is not possible to relate in detail the various steps in 
the progress, but it will 
readily be understood that 
very considerable work and 
expenditure were involved in 
the experimental develop- 
ment and manufacture of the 
lamps which were finally de- 
cided upon as satisfactory for 

use. A new condition had to be met. Life in lamps hitherto had 
been measured in hours of incandescence without regard to frequency 
of illumination. For switchboard use, life had rather to be measured 
by the number of times that the circuit could be made and broken 
without the utility of the filament being impaired. 

The drawings figs. 136-139 from British specification No. 11,549 
of 1895 illustrate the lamp as first put into practical use in the com- 
mon battery switchboard. Fig. 136 is a section drawing of the lamp , 

FIG. 137. 

fig. 137 shows the lamp inserted into the switchboard and connected 
to the contact springs, and fig. 138 a part of the ebonite strip in 
which the lamps were mounted. An important detail in this lamp 
is the plano-convex lens of opal glass (fig. 139). 

This lens serves two purposes : one is to disperse the light from 
the lamp so that the signal can be seen from the side as well as 
from the front ; the other is to prevent the red glow of the filament, 
when both lamps are in parallel, from being seen and mistaken for 
a signal. 1 

1 British specification, No. 11,549, 1895, P- 9i to ne 3 T - 

The second purpose applied only to one of the various ways 
of connecting the supervisory signals and was not adopted in practice, 
but the illumination to the side was of great importance and has 
been used continuously. At night or slack periods it is necessary 
that an operator should be able to see the indication of a call at 
some distance to right or left of the indicator. The plano-convex 
lens provided for this condition in a very simple manner without 
detracting in any way from the utility of the signal to an operator 
immediately in front. 

In the committee of May 1895 Mr. Carty said : 

Mr. Scribner exhibited a lamp not long ago which was provided 
with an opalescent shade. I witnessed the operation of that in 
a model, and it seemed to come near the ideal. 

No doubt the model referred to was the plano-convex lens, for 
which a British patent was applied for in the month following. 
The relationship of the indicator to the jack was always a matter 

FIG. 139. 

of importance. In some of the earlier switchboards there was great 
advantage in that the jack was placed close to the indicator of its 
own line. In the standard and multiple board the balance of 
advantage was found in separating them. The construction of an 
indicator of the lamp form occupying no more space than a spring 
jack, permitted the arrangement of indicators and jacks in such close 
relationship that the insertion of a plug in the jack appropriate to 
a glowing lamp could be carried out with a minimum of conscious 
effort on the part of the operator. 

The lamps were mounted ten on a strip and the answering jacks 
also, it being preferred, as the specification states (p. 9, line 39), 

to place each line lamp over the corresponding answering jack. The 
two rlearing-out signals belonging to a pair of plugs are placed in the 
key-shelf in front of and in line with the plugs which they represent. 

An arrangement which has not needed to be altered, except that 
the lamps are under instead of over the spring jacks. 

The tongs shown in Fig. 140 are to enable a lamp to be ' quickly 
removed and replaced by another in case of failure of the lamp.' 

The use of a lamp as the indicator of a telephone call was a very 


great advance. 1 In its primary purpose of attracting attention it 
was far superior to the shutter forms previously in use. Whilst 
the shutter indicator needed watching, the lamp seemed to speak 
for itself, and the announcement it was impossible to ignore. A 
secondary advantage lay in being able to remove the operative 
mechanism from the face of the switchboard. In the indicator it 
was not possible to separate the shutter from its electro-magnet, but 
when the lamp was substituted for the shutter, the relay which was 
the equivalent of the electro-magnet could be placed at a distance, 
thus leaving valuable space to be occupied by apparatus which the 
operator needed to handle. 

In a preference for lamps as signals the members of the 1895 
committee were unanimous. There were differences of opinion as 
to the method by which the lamps should be operated with or 
without relays. The signal lamps used in Chicago were controlled 
by relays, the lamps being in the local circuits. Regarding them 
Mr. Scribner said : 

If the hook is moved too rapidly, the relay fails to open and 
close the lamp circuit. 

You can hear the vibration S^ :z: ^\ 

of the relay at the board, \ f 

but the lamp does not 

respond. C 

As illustrating a pre- FIG. 140. 

vailing sentiment at the 

time, I continue the quotation from Mr. Carty's remarks given 
on p. 384. 

Now if we could have such a lamp as that operated directly 
without any relay, I would like it. If it could not be done I think 
a lamp like that operated by an electro-magnet would be satis- 
factory. I would prefer a lamp to a ' magnetic signal.' 

A lamp operated directly without any relay was thus regarded 
as promising for the supervisory signals. There was less objection 
to the relay for the line because the action would be direct and the 
responsiveness to quick changes would not be required. A method 
including this combination is illustrated in British specification, 
No. 11,549, J une i%95> already referred to. 

1 ' The prime advantages of lamp signals are that they are extremely 
compact, they have no working parts, and therefore may be placed in any 
position, vertical, horizontal, or at an angle ; they are automatic in action 
or self-effacing, since the signal disappears immediately the current is cut 
off ; and finally they give a much more positive and assertive signal than any 
form of indicator.' H. L. Webb, Journal of the Institution of Electrical 
Engineers, May n, 1905, xxxv. 291. 

2 C 


It was necessary that the supervisory signals should be ' positive 
in their action ' ; that is, ' that they should by their condition indicate 
unequivocally to the operator either that the subscribers' telephones 
were on the hooks or were off the hooks.' 1 The supervisory signals 
were needed to be responsive to that fluttering of the hook by which 
subscribers were able to attract an operator's attention to a con- 
nected line, and so far relays were not responsive enough for this 
purpose. A new design was required. The design which in the 
main principle met this requirement is also included in No. 11,549 f 
1895. I am not familiar with all the varieties of all the relays which 
have been invented for telegraphic use. Their number must be 
very large, but I am probably safe in saying that the armature which 
was attracted by the magnet had a hinge of some sort. Now the 
relay which was devised expressly for the common battery system 
had an important share in making that system practicable, and 
requires therefore some attention. 

The illustration fig. 141 is the relay included in fig. 2 of No. 11,549 
of 1895. It is extremely unlikely that Bell's 
experiments of twenty years before were in 
the mind of its designer, but in comparing 
this relay with its predecessors, one is 
irresistibly reminded of Bell's experiments 
with armatures in 1875. 2 Armatures with 
hinges or equivalent methods of support 

W6r6 discarded ' and the disc was attached 
directly to the diaphragm. Similarly in the 

design of a relay which should be readily 

responsive it was found advisable to discard any support fixed to 
the armature. The description in the patent is very brief : 

The relay shown in fig. [141] is a tubular magnet with a disc 
armature which is brought to an edge at its periphery, the disc 
resting on this edge in an annular groove formed between the end 
of the tubular magnet and a shoulder in the cap. When the 
armature is attracted it closes against an insulated contact stud 
which projects from the magnet core. When not attracted it 
drops away by gravity from this contact. 3 

The tubular form of this relay also recalls the iron box receiver 
exhibited at the Centennial, 4 though the method of operation was 
different. Magnetic attraction on the one hand, gravity on the 
other, were not seriously impeded by friction. The armature rested 
upon the tube, and the surface at the point of support was reduced 
to the smallest possible dimensions by being ' brought to an edge 
at its periphery.' 

1 Carty, Herzog Case, p. 33. * Chapter vi. p. 48. 

' P. 9, line 49 of specification. * Chapter vi. p. 51. 



The movement of the armature away from the electro-magnet 
was restricted by the ' annular groove.' It is clear that such 
restriction could be obtained by other means. It is also clear that 
the circular or disc formation of the armature was only useful with 
a ' tubular magnet.' In subsequent modifications the tubular form 
of magnet and the circular armature were abandoned, but the base 
of the latter was still ' brought to an edge ' and became known as 
a ' knife edge.' It was the ' knife edge ' which gave its name to this 
form of relay. 

The inventor of the knife-edge relay was Mark A. Edson of 
Chicago. His United States patent specification 1 contemplated 
the 1 combination of a relay and incandescent lamp in ' a single 
appliance or instrumentality.' The illustration fig. 142 is fig. i 
of his specification : 

The original description of the operation and statement of the 
advantages of his form of armature are even clearer and more 

FIG. 142. Edson Knife-edge Relay. 

definite than those used in respect to the apparatus which was 
developed from it, and quoted above from the British specification 
No. 11,549 of 1895. 'The original inventor says : 

9 is a movable tilting soft iron disc armature shaped as a 
truncated cone, which rests upon the floor of the cap 6 by the edge 
of its base, its lower inner surface bearing upon the end of the metal 
tube 10, and being held in place and made adjustable by the adjusting 
screw 8. A great advantage of this construction is that the arma- 
ture resting on its sharp edge on the internal periphery of a circle 
considerably larger than itself is almost frictionless. 2 

It is probable that the confidence inspired in the improved 
relay led to the modification of the circuits and the adoption of 
relays for controlling all the lamps. Thus the common battery 
line circuit within the central office approximated to that of the 

1 No. 550,260, dated November 26, 1895 (application filed May 20, 1895). 

2 U.S. specification, No. 550,260, p. i, line 88. 

2 c 2 


branching system, where an indicator consisted of two coils known 
respectively as the line coil and restoring coil. A magneto-current 
sent over the line dropped a shutter, and the insertion of a plug in 
the jack of the line by an operator restored the shutter. By 
analogy the line relay and the cut-off relay of the common battery 
were the equivalents of the line and restoring coils of the branching 
board. A current coming over the line energises the line relay 
which lights the line lamp. The insertion of a plug in the 
appropriate jack energises the cut-off relay which breaks the circuit 
of the line relay and extinguishes the line lamp. The analogy of 
the line and cut-off relays with the coils of the branching indicator 
is so far carried into the construction that the line relay and the 
cut-off relay are mounted together, though not part of the same 
structure, as were the line and restoring coils of the branching 

The feature of the common battery system which commended 
itself so strongly to telephone engineers was the improvement in 
the service which was rendered possible by increasing the number 
of signals in a pair of connected lines from one to two, 1 by rendering 
the normal operation of those signals automatic, yet providing a 
means of attracting the operator's attention by either subscriber 
before disconnection. The last-mentioned operation is effected 
by the alternating movement of the telephone hook at the sub- 
station, and it becomes essential that the relay should respond to 
the movements, however rapid they may be, so that the lamp may 
reflect the movement of the hook. The knife-edge relay was equal 
to the demand upon it, and the supervisory lamps were accordingly 
controlled by relays as well as the line lamps. The development 
of suitable relays was an important feature of the common battery 
system ; in fact, the technical term descriptive of a full common 
battery equipment was the ' No. I Relay Board.' 

Following the experimental plants previously recorded, the 
first large commercial installation of the common battery system 
was at Worcester, Mass., in June 1896. It was designed for employ- 
ment in connection with common battery transmission, but when 
first installed the common battery was used for signalling only, 
local batteries being used for talking. A relay was employed for 
the line signal, this line relay served also for the supervisory signals. 

1 ' In my opinion the real dividing line between the periods mentioned is 
the introduction of double supervisory signals. . . . The weakness of the 
ring off indicator was not due to the fact that it failed to work when required, 
. . . but . . . that it [was] worked by the subscriber who had to perform 
a deliberate act that had no necessary connection with the work in hand ; . . . 
it did not give continuous supervision . . . and . . . independent know- 
ledge of each side of the circuit.' W. W. Cook, Journal of the Institution oj 
Electrical Engineers, May n, 1905, xxxv. 304. 





A circuit of the supervisory lamp, which was completed when a 
plug was inserted in the spring jack, served to efface the line signal 
by virtue of a shunt circuit established around the line lamp by 
the supervisory lamp. Hanging the telephone upon the switch 
interrupted the current of the line relay which de-energized and 
released its armature, thus opening the contact previously made 
and interrupted the circuit through the line lamp. This served 
to increase the current through the supervisory lamp to an extent 
sufficient to light it. 1 

The Worcester installation was the public experimental debut 
of the common battery system. Following the method by this 
time well established, a new system was devised in the laboratory 
tested on a sufficiently practical scale in a private exchange, and, 
proving successful under those tests, was put into practical use in 
a public exchange. The Worcester essay was successful, but 
showed that modifications might be effected. The modifications 
were tried in private and were included in the next public exchange 
at Louisville, opened in September 1897. In the Louisville outfit 
were to be found the cut-off relay and supervisory relays to control 
the supervisory lamps. 

In these switchboards a common battery was employed for 
operating the line relays and the supervisory relays, and for supplying 
current to the subscribers' transmitters over their lines. A second 
common battery of a less voltage was provided for supplying the 
supervisory lamp with current ; and still another common battery 
of still lower voltage for supplying the operators' transmitters and 
the line lamps with current, and the system has been termed the 
three voltage relay system. 2 

A few other central offices installed early in 1898 were of the 
three voltage type. Those installed later in the same year were of 
the single voltage type in which all of the apparatus of the equip- 
ment was operated from a single common battery. 2 

The Harlem (N.Y.) board, which was under discussion in the 
committee of 1895, was brought into service November 12, 1898. 
The plans for this board were made with due regard to the require- 
ments of the whole New York service, and the results of its operation 
were such as to determine the complete conversion of the New 
York system from the magneto branching to the common battery 
system. Four central offices were completed in 1899, three in 1900, 
four in 1901, three in 1902, and two in 1903. These included equip- 
ment for 73,400 lines. Descriptive articles appeared in the American 
electrical press from time to time during this period. Other cities 
in the United States were also being fitted at the same time. 

1 Scribner, Herzog Case, p. 555. 2 Ibid. p. 557. 


The first installation in England was at Bristol, completed in 
1900 ; the next at the London Wall exchange in London. When Mr. 
Gill became engineer-in-chief of the National Telephone Company, 
the common battery system was adopted generally in the principal 
central offices of that company. Sir John Gavey decided on the 
adoption of the common battery system for the Post Office service 
in London. The first central office in Carter Lane, E.G., was 
opened in April 1902. 

From a description of the Post Office system published in The 


Side Elevation 

Front Elevation 

Plan of Lower (Cut-off) Relay 
FIG. 145. Line and Cut-off Relay. 

Electrician* figs. 143, 144, and 145 are taken. They clearly 
illustrate the circuits. 

The iUustration (fig. 145) may be compared with fig. 141 
to show the development from the disc armature. 

When first introduced the jacks were larger than is now general, 
and the lamps were mounted ten on a strip. Later it was found 
possible to manufacture lamps sufficiently small to be mounted 
twenty on a strip of the same length as a jack strip with a similar 
number of spring jacks. 

1 March 14, 1902, xlviii. 808-10. 



THE switchboards so far described are elaborate machines, conveying 
certain information by what may be called automatic means but 
depending upon the service of an attendant for the connecting 
together of any two subscribers' lines. 

We have now to consider even more elaborate machines whereby 
the position is to some extent reversed, the subscriber performing 
certain definite and conscious operations which, by means of suitable 
machinery, effect the connections desired. The machines have become 
conventionally known as automatic switchboards, and though the 
accuracy of the term is questioned, its adoption has become so general 
that it would be inconvenient at the present stage to use any other. 

Telegraphic instruments sending signals without an operator 
have been called automatic telegraphs, as for example ' Wheat- 
stone's Automatic,' though the paper slip used therein is previously 
punched by hand. Such systems have of late years been frequently 
called machine telegraphy. Perhaps the automatic telephone system, 
if it survives, will undergo a similar change. ' Automatic ' is suffici- 
ently accurate to describe the switchboard, which is really an automa- 
ton. It becomes inaccurate and somewhat misleading when applied 
to the system as a whole, for the so-called automatic system requires 
conscious effort on the part of the subscriber for every stage of opera- 
tion, whilst much of the so-called manual system is carried out auto- 
matically without any conscious effort on the part of the subscriber. 

One of the earliest examples of the machine or automatic type 
of switchboard, and the earliest in patent date, is that referred to in 
the following letter : 

Law Offices of Connolly Brothers, 

Patents and Patent Causes, 

September 17, 1884. 

DEAR SIR, We tender you and the members of the National 
Telephone Association a cordial invitation to inspect, criticise, and 



improve upon our Automatic Telephone Switch, which will be set 
up for exhibition to-day at the Electrical Exposition. 

As this invention has been productive of much controversy 
among telephone people, many of whom asserted it was impossible 
to effect what it proposed, and as it does its work, enabling subscribers 
to effect connection of their lines interchangeably, without personal 
or manual service at the Central Office, we feel confident that an 
inspection of it will be interesting to yourself and the members 
of the Association. 

The location of the exhibit is in Section U 15, near B. & O. 
Telegraph Office, S.E. corner of Exposition Building. 

Very respectfully, 


Morris F. Tyler, Esq., President, 
National Telegraph Exchange. 1 

The Connolly Automatic Patent is U.S. No. 222,458, dated 
December 9, i8jg. 2 It is the joint invention of Messrs. M. D. and 
T. A. Connolly of Philadelphia and Thomas J. McTighe of Pittsburg, 
and is entitled ' Improvement in Automatic Telephone Exchanges,' 
suggesting that automatic exchanges of some kind were already in 
existence. The patentees say : 

The object of our invention is to provide what may be termed 
an ' automatic telephonic exchange ' in which each station of the 
exchange is in communication with a main or principal station, 
through which connections are established between any two of the 
individual stations. 3 

Conforming to the usual practice in the preamble of patent 
specifications, the patentees relate the prior method and indicate 
their own view of the unsatisfactory results. 

Under the present system in use in the principal cities having 
telephonic facilities, the lines from the several stations converge 
to a central office and terminate in a switchboard. When any 
individual member of the exchange desires to communicate with 
any other member he signals the central office, states his desires, 
and an attendant thereupon makes the desired connection. The 
operation of making these connections is now altogether a manual 
work, and requires not only constant attention but much dexterity 
in order that there shall be as little delay as possible ; but in 
exchanges comprising many members, the work of the central office 
is very great, requiring many employes to meet the wants of the 
community. Even then there are incessant delays, much confusion, 

1 National Telephone Exchange Association Report, 1884, p. 76. 

2 Application filed September 10, 1879. 

8 U.S. specification, No. 222,458, December 9, 1879 (application filed 
September 10, 1879). 



and consequently many mistakes and annoyances which it is highly 
important should be obviated. 1 

The means to be adopted to obviate all these mistakes and 
annoyances very naturally follow : 

Our present invention contemplates the employment, in lieu 
of manual labour and the necessary skill and intelligence to apply 
it, of the capabilities of electricity and electro-magnetism whereby 
all the difficulties now met with are entirely overcome and the 
operation of the central office rendered completely automatic, 
rapid, and reliable. Each or any member of the exchange may, 
by means of local contrivances having electrical communication 
with the central office, place himself in communication with any 
other member whose line happens to be unoccupied. At the same 

FIG. 146. Connolly Auto- FIG. 147. Froment Telegraph Trans- 

matic (Dial). mitting Dial. 

time he is enabled to entirely isolate his own and the line he desires 
to communicate with from all others of the exchange so that no 
interferences or interruptions can possibly occur. He is also 
enabled to signal the member to be communicated with, and in 
fact to place his own and the other line into the most desirable 
and convenient relation to each other and to the balance of the 
exchange as the most urgent demands of the telephonic-exchange 
system require. 1 

Considering the date of the invention, there are features of much 
interest in Connolly and McTighe's specification, but space will not 
permit extended reference to its mechanical details. In automatic 
apparatus it is natural to turn first to their limitations. ' Upon the 
face of the dial are indicated the numbers or letters of the different 

1 U.S. specification, No. 222,458, December g, 1879 (application filed 
September 10, 1879). 


stations in the system.' The relation of the pointer to the teeth on 
the periphery is described. 

These teeth correspond in number and position with the number 
or letters on the dial, so that when the pointer coincides with, say, 
100 on the dial, it indicates that 100 teeth have passed the spring 
and a like number of breaks have been made. 1 

While 100 is mentioned in the letterpress description, the 
dials shown in the drawings have no numbers exceeding 25, being in 

FIG. 148. Connolly Automatic (Central Office). 

this respect no advance on pulsation-transmitting devices previously 
used in dial telegraphs. Pouillet's ' Elements de Physique Ex- 
perimentale/ seventh edition (1856), 2 describes the ordinary dial 
telegraph and particular improvements effected therein by certain 
specified inventors, including those of Froment presented to the 
Academy of Sciences in 1851. 

Comparison of fig. 146 enlarged from Connolly's specification 
with fig. 147 from Pouillet will show that the selecting apparatus 

1 U.S. specification, No. 222,458, December 9, 1879 (application filed 
September 10, 1879). 2 Vol. i. 340. 



of Connolly was practically the well-known dial telegraph which 
had been in existence for about thirty years. 

The similarity of operation at the receiving station will also be 
observed from the comparison of Connolly's illustration indicating 
the central office (fig. 148) and that of Pouillet illustrating the re- 
ceiving apparatus of an ordinary dial telegraph (fig. 149). Froment's 
was described in detail by Shaffner 1 and the manipulator and 
receiver illustrated. 

The twenty-five detents in Connolly and McTighe's illustrations 
are probably due to the fact that they took as a model an instru- 
ment intended to indicate all the letters of the alphabet less one 
the W was omitted by reason of the French origin of this invention. 
They describe its operation and say : 

This is to be understood as 
merely a suggestion as to the 
means of breaking the circuit 
and indicating the intermis- 
sions. Any of the well-known 
forms of dial instruments 
adapted to the use required 
may be employed, and hence 
our invention is not limited to 
any particular one. 2 

The novelty of the inven- 
tion begins with the multipli- 
cation of the receiving ratchet 
wheels E, there being one for 
each station and for each 
ratchet a pointer travelling 

with it, and a fixed segment G. Beyond the segment there is for 
each line a bar I long enough to reach from end to end of the series 
of segments G and arranged ' parallel with the common axis ' 
(i.e. shown in cross-section in fig. 148). From each bar project 
hooks i l of metal, 

there being as many hooks on each bar as there are ratchets and 
pointers i.e. one hook for each circuit that enters the central office 
and these hooks are made to stand normally, in such position 
that the hook e l of the slide e on pointer E 2 will pass through, and 
in electrical contact with, hook i 1 when the pointer is rotated.* 

The contact between the pointer and the hook of bar I is made 

1 The Telegraph Manual, 1859, p. 374. 

* U.S. specification, No. 222,458, December 9, 1879 (application filed 
September 10, 1879). 

FIG. 149. Froment Telegraph 
Receiving Apparatus. 


by a sleeve e with a longitudinal movement on the pointer arm. 
The sleeve e has a projection with a right angle bend e l which 
engages with another right angle bend i 1 on bar I. The reversing 
key is an important feature of the apparatus. In the words of 
the patentees it ' effects several remarkable results.' One of these 
is that the magnet K has its poles changed, attracts the bar I 
' which is pivoted at i 2 (or otherwise arranged so as to be moved 
radially from the centre) and pulls the latter out.' The effect 
cannot be more graphically or succinctly described than in the 
language of the patentees : 

The result is that the hooks i 1 of bar I are now out of the path 
of the hooks of all other pointers, and consequently the circuit thus 
established between the two local stations is completely isolated, 
and the pointer of the central office ratchet of no other local station 
can obtain a contact with said circuit. Hence absolute immunity 
against interruption, and all the annoyances of cutting in and 
cutting out, cross-talking, etc., is afforded, thus ensuring the utmost 
privacy for the stations wishing to converse. 1 

The work of the subscribers is thus described : 

To sum up, then, the operations required are as follows : At 
the station desiring to call another station, operator [i.e. subscriber] 
whirls round his index till it arrives opposite the number of the 
station desired, reverses his key and then sets his switch on the 
call-bell. As soon as the answering signal is received, he sets his 
switch to the third point, which shunts the local battery into the 
primary of the telephone and places the secondary in the main line. 
After he is done talking, he switches on the battery to the main 
line and places his reversing key back to normal, which restores the 
central office to normal, and then he whirls his index around to 
zero and switches his bell into main. 

At the station called, as soon as the bell rings the operator 
switches his battery into the line with its direction conforming to 
that required to preserve the armatures at central office in the 
positions set by station calling. This will at once ring the bell of 
the station calling, after which he switches his telephone secondary 
to the line and battery to the telephone. After the connection is 
finished he switches back to the bell simply. 1 

The opinion of the inventors in regard to the capabilities of the 
average telephone user was evidently high, for they say : 

These are simple operations, requiring no skill, and may be all 
performed successively within the period of a moment or two. 1 

1 U.S. specification, No. 222,458, December 9, 1879 (application filed 
September 10, 1879). 


In modern automatic phraseology their invention may be said 
to combine the features of a ' connector ' and a ' selector,' a point 
of much interest being the method by which a connected ' connector ' 
was removed from the path of other ' selectors.' 

The Connolly system was exhibited at the Electrical Exhibition 
in Paris in 1881, and was described in the Journal Telegraphique of 
December 25, 1881, by M. Rothen. 1 The installation was for eight 
stations. The mechanism, says M. Rothen, was certainly very 
ingenious and worked well, but would it do the same in working 
on real lines ? He could not disguise from himself that success 
was very doubtful. It was perhaps M. Rothen's doubts of 1881 
that Messrs. Connolly referred to in their letter of 1884 at the 
beginning of this chapter, and that prompted them to call the 
attention of the members of the telephone convention to the fact 
that ' it does its work.' 

About a month later than the Connolly and McTighe patent date, 
George Westinghouse (junior) of Pittsburg, Pennsylvania, applied 
for patents on automatic exchange apparatus. 2 In these applica- 
tions, Westinghouse did not contemplate the adoption of auto- 
matic working in a large exchange, but provided only for its use 
as an auxiliary in outlying areas. After describing the ordinary 
use of the telephone he says : 

It has also been found that in suburban or outlying villages, 
boroughs, etc., a few miles distant from the central exchange, a 
few persons frequently reside who desire to be in telephonic com- 
munication at their residences with one or more of the users of 
telephones in the city, but the number of such suburban residents 
in any one locality is frequently so small that it does not pay com- 
mercially to maintain a separate local exchange for them, and the 
distance is so great that the cost of a separate wire for each, leading 
to the main city exchange, prevents them from enjoying the desired 
advantages and conveniences of a home telephone. . . . 

By my present invention I enable each such suburban user to 
call and open telephonic communication through the auxiliary 
exchange with the main or city exchange, and through it with any 
desired city user without the necessary intervention of an operator 
at the auxiliary exchange. In doing this he automatically locks 
the opening or closing connections of the other suburban users 
having connections with the same auxiliary exchange, so that 
they cannot call or interfere with his use of the line until he is 

The invention also includes provision by which, when such user 
is through [i.e. completed his conversation], the operator at the main 

1 Quoted by Du Moncel, Le Tiliphone, fourth (French) edition, p. 365. 

2 U.S. specifications, Nos. 223,201, 223,202, dated December 30, 1879 
(applications filed respectively October n and 13, 1879). 


or city exchange can restore the apparatus at the auxiliary exchange 
to its normal condition, so that any other suburban user can call 
and hold telephonic communication in like manner with a city user, 
and also in so doing lock out his suburban co-users as before. 1 

Westinghouse contemplated an auxiliary automatic exchange, 
and the system suffered from the same limitations as did the party 
line the willingness of the subscribers to participate in a service 
in which they were liable to be locked out by their co-users. 

Inventors in the field of automatic switchboards in the main 
limited themselves to the provision of apparatus serving the purpose 
so clearly described by Westinghouse in 1879. 2 Mr. D. Sinclair 
installed at Glasgow some experimental apparatus in 1883. 3 

The Electrical Review of London, November 17, i883, 4 describes 
a system the English rights of which had been acquired by Mr. A. S. 
Paul. This, according to my recollection, was the production of 
Ericsson of Stockholm, and was, like Westinghouse's, intended only 
for auxiliary exchanges. It worked like Connolly's, by a step-by- 
step arrangement. There were numerous other inventions of the 
same class none being adapted for use by large numbers of sub- 
scribers. Mr. Kempster Miller 5 considers that the first steps towards 
the realisation of the ideas advanced by Connolly and McTighe were 
the inventions made by Strowger, commencing with his U.S. patent 
No. 447, 918, dated March 10, 1891. This application was filed on 
March 12, 1889, approximately ten years after that of Connolly and 
McTighe. While an advance upon its predecessor in the more elab- 
orate operation of the central office connecting device, it must be 
considered as retrograde in that, instead of the single wire of Connolly, 
it required five wires from the sub-station to the central station. 

Fig. 150 is a reproduction of fig. I of Strowger's specification, 
No. 447,918. The method of operation is thus described in the 
specification : 

The person wishing to place his transmitter and ear-phone in 
connection with those of another, he will do so by successively 
pressing or depressing the keys, which cause the circuit closer 
C C' [a pointer or connecting arm revolving within the cylinder 
A and thus not shown in the diagram] to move. For example, 
if telephone 288 wishes to place himself in connection with telephone 
315, he will do so by pressing the key marked G' three times, then 
the key marked H' once, and then the key marked I' five times. 
His circuit closer C C' is then in contact with wire terminal 

1 U.S. specification,